Implications for Lunar Volcanism

from Studies of Cryptomafic Deposits





Irene Antonenko




Department of Geological Sciences, Brown University
Providence, Rhode Island 02912 USA







Table of Contents



Abstract
       Studies of cryptomafic deposits have important implications for the total volcanic flux and output on the Moon. In order to better understand global lunar volcanism, a survey of suggested cryptomafic deposits was conducted. Estimates of volumes and areas were determined for each known cryptomafic deposit located on the Moon, and their age roughly calculated. Many cryptomafic deposits may still be undiscovered, thus these areal and volume estimates should not be considered as comprehensive. Regardless, the results of this study have shown that cryptomafic deposits on the Moon may be extremely significant, potentially representing a total area of as much as 3.6 x 106 km2 (~50% of known mare area) and volume of 1.6 x 106 km3. Mare volumes were recalculated to be ~5 x 106 km3, thus cryptomafic materials may represent ~30% of known mare volumes. Extrusive volcanism would, therefore, represent a total area of 9.5 x 106 km2 (25% of the surface area of the Moon) and a total volume of 6.6 x 106 km3. The fluxes of early (>3.8 b.y. old) volcanics are suggested to be ~4.5 x 10-3 km3/a, or 60% of the flux in the late Imbrian period (3.8-3.2 b.y.). The implications of these results is that volcanism may have begun much sooner than had previously been recognized and was much more voluminous during the early period of lunar history. Such findings would indicate that the early lunar interior was hotter than previously believed and that extensive melting of source regions must have begun at an earlier time. These results may place important constraints on models of lunar evolution, which should be re-evaluated with these constraints in mind.



Introduction

       Mare deposits on the Moon represent the extrusion of liquid magmas onto the lunar surface. Their volumes and ages provide important data points in the determination of the size and distribution of magma source regions [Yingst and Head, 1994], the timing and sequence of basalt production and extrusion [Hess and Parmentier, 1995], and the thermal history and evolution of the Moon [Solomon and Head, 1979, 1980; Solomon et al., 1981; Head and Wilson, 1992]. To address these issues, it is important to have a complete understanding of all lunar volcanism. However, the identification of cryptomafic or hidden volcanic deposits [Schultz and Spudis, 1979, 1983; Hawke and Bell, 1981; Bell and Hawke, 1984; Mustard et al., 1992, 1994; Head and Wilson, 1992; Antonenko et al., 1995] has brought our understanding of lunar volcanism into question.
       Originally, the age span of mare basalts samples, returned from the Apollo missions to the Moon, was thought to be approximately 3.9 - 2.5 b.y., corresponding to the ages of the observed mare deposits [Head, 1976]. It was noted that this entire range of ages could often be found in a single basin [Head, 1976], indicating that lunar volcanism was occurring over an extended period of time. Some authors even suggested the existence of a minor phase of earlier volcanism, which predated the known mare deposits [Taylor, 1975]. This hypothesis was substantiated when ancient basalt clasts, some as old as 4.36 b.y. old, were found in the Apollo samples [i.e., Taylor et al., 1983]. These ages do not correspond to the ages of any of the known lunar mare deposits [Head, 1976], thus suggesting that they may originate from ancient cryptomafic deposits.
       In order to evaluate the global importance of cryptomafic deposits, a census of all cryptomafic materials on the Moon would be required, where the location, extent, and volumetric contribution of each deposit was documented in great detail. Such a program of study, however, would require many years to complete and is well beyond the scope of this thesis. Nevertheless, it is possible to estimate the total impact of cryptomafic deposits on lunar volcanism by considering those cryptomafic deposits that have been previously documented. To this end, a literature survey was conducted to catalog all previously identified and suspected cryptomafic deposits. The locations of possible cryptomafic areas, thus determined, are listed in Table 1 and illustrated in Figure 1. It is not suggested that this list is comprehensive; in all probability, many potential cryptomafic areas still remain unidentified and will require careful processing and study of the data to be discovered. However, this list provides a good beginning for the study of global cryptomafic contributions.
       In the previous chapters, several techniques for determining the volumes of cryptomafic units were described. For accurate volume estimates, a detailed analysis, using the above techniques, would be desirable for each of the cryptomafic areas identified in Table 1. Again, such a detailed analysis is well beyond the scope of this thesis. However, rough estimates of the volumes of cryptomafic deposits can be calculated for each area on the basis of information in the published literature and the experiences gained during previous studies (Chapters 1-3).
       Thus motivated, a cursory survey of all known or suspected cryptomafic deposits was begun (Table 1), using previously published findings and a limited analysis of existing data. Rough estimates of cryptomafic volumes were obtained for each area and the possible ages of these deposits were determined. These findings were then evaluated, in order to assess how they affect our understanding of global lunar volcanism and lunar evolution models.



Global Estimates of Cryptomafic Deposits

       Cryptomafic deposits can be identified by a number of criteria [Antonenko et al., 1995], and many of these criteria can also be used to help determine the boundaries of these deposits. For example, the presence and distribution of dark-haloed craters, the extent of light plains units, and topographic boundaries have all been used in Chapter 1 to help in boundary delimitation.
       Schultz and Spudis [1979] conducted a global survey of all lunar dark-haloed impact craters and compiled their results in a map (Figure 2). Because dark-haloed craters are often associated with cryptomafic materials, the distribution of these craters in target areas can be used to identify the boundaries of cryptomafic deposits, as was done in Chapter 1. However, the data from the Schultz and Spudis [1979] survey has several limitations. At the time this survey was conducted, the existence of basalt-excavating craters was still in question and the authors only catalogued the largest and most convincing craters in order to conclusively demonstrate that basalt excavation was indeed occurring [Schultz and Spudis, 1979]. Many possible dark-haloed craters are, therefore, not included on their map (Figure 2). Thus, the absence of documented dark-haloed craters does not provide conclusive evidence for the absence of a cryptomafic deposit and the boundaries of cryptomafic deposits may extend beyond the immediate areas surrounding known dark-haloed craters. Thus, when dark-haloed craters are present, they can provide a rough minimum estimate of the boundaries of cryptomafic deposits.
       It has long been suspected that many light plains deposits (smooth regions of intermediate albedo) may hide cryptomafic materials [i.e., Bell and Hawke, 1984]. Thus, when light plains units are clearly identified as obscuring cryptomafic materials, the extent of these units can be used as an aid in estimating the boundaries of the cryptomafic deposits. Deposits of light plains-type units are well documented on lunar maps [i.e., Lucchitta, 1978; Wilhelms and El-Baz, 1977; Stuart-Alexander, 1978] and Howard et al. [1974] have compiled a global light plains map (Figure 3). These maps can be used at various target locations to help determine boundaries of cryptomafic deposits, when appropriate. However, cryptomafic materials may also occur outside of light plains units [i.e., Mustard and Head, 1996], thus light plains deposits, when they are associated with cryptomafic materials, provide only minimum estimates for the boundaries of these cryptomafic units. Also, many light plains units have been shown to consist of Orientale or Imbrium ejecta deposits [i.e., Chao et al., 1973]. When the origin of a mafic-obscuring ejecta deposit is known, it can provide an upper bound for the age of the underlying unit, helping to date the cryptomafic deposit.
       Mare deposits are found to be preferentially located in basins [Head, 1975a], corresponding to areas of crustal thinning [Head and Wilson, 1992; Zuber et al., 1994]. The presence of a basin or a region of crustal thinning can, therefore, be used to help determine the extent of a cryptomafic unit. The Clementine gravity data set provides a good opportunity for the identification of such features. Basins, topographic lows, and regions of crustal thinning can be identified on maps of lunar topography (Figure 4a) and crustal thickness (Figure 4b), presented by Zuber et al. [1994] and Neumann et al. [1996]. Refinements to the crustal thickness map have been performed by Wieczorek and Philips [1998], however, their findings are more or less consistent with earlier results. Volcanic deposits are not necessarily defined by the basins that contain them. For example, Mare Humorum is confined to the inner ring of Humorum basin [Wilhelms, 1987], while the Humorum cryptomafic deposit (Chapter 1) spreads out beyond the outer ring of Humorum basin. Thus, basin topography and crustal thinning can only give very rough estimates of cryptomafic boundaries.
       Geochemical data has also been used to identify the presence and extent of cryptomafic bodies [i.e., Maxwell and Andre, 1981; Hawke et al., 1985]. Here again, the Clementine data provides an excellent tool for identifying chemical anomalies on the lunar surface. Lucey et al. [1995,1998a] and Blewett et al. [1997] have compiled a map of FeO abundance on the Moon (Figure 5), using Clementine UVVIS data. Anomalously high FeO values, outside of identified mare areas, can potentially be associated with cryptomafic deposits [Antonenko et al., 1995]. The FeO abundance map of Lucey et al. [1995,1998a] and Blewett et al. [1997] (Figure 5) may, therefore, be used to help identify the extent of cryptomafic areas that have distinguishable FeO anomalies. Since geochemical anomalies are generally associated only with thinly covered cryptomafic deposits [Antonenko et al., 1995], not all cryptomafic bodies can be identified this way. Also, FeO anomalies will tend to underestimate the boundaries of cryptomafic deposits, since the anomalies may disappear in areas where thicker deposits cover the cryptomafic material.
       Both light plains deposits and cryptomafic units are associated with albedos that are intermediate between that of highland and mare materials [Head, 1972; Antonenko et al., 1995]. Changes in albedo can thus be useful in determining the placement of cryptomafic boundaries. The Clementine albedo map (Figure 6), prepared by the USGS [1997], provides a very useful tool for identifying albedo anomalies. It should be noted that young, bright craters in the vicinity of a cryptomafic deposit may significantly raise the albedo of sections, or all, of the cryptomafic unit [i.e., Maxwell and El-Baz, 1978; Hawke and Spudis, 1980; Hawke et al., 1985; 1993], thereby giving only a minimum estimate for the boundary of the cryptomafic unit.
       Using as wide a variety of the above data sets as possible, as well as relying on other published information and my experience in determining cryptomafic boundaries, the potential extent of each cryptomafic deposit, identified in Table 1 and Figure 1, is determined. From these, rough areal estimates are calculated for each cryptomafic region. Since most of the methods outlined above give minimum estimates of cryptomafic boundaries, these areal estimates will tend to be conservative. At each cryptomafic location, a low estimate of thickness is determined for the cryptomafic body on the basis of published data for known mare thicknesses in nearby regions. From these values, the possible volume of cryptomafic material at each location is roughly determined. Again, because the area and thickness estimates used in obtaining these volumes are extremely conservative, it is believed that these are minimum volume estimates. An approximate estimate of the age range for each cryptomafic region is also determined on the basis of known stratigraphic relationships. The findings from this study are listed in Table 1.


        Western Limb Region
       Cryptomafic areas in the western limb region of the Moon, namely the Schiller-Schickard area, areas west of Humorum basin, and areas west of Oceanus Procellarum, have been studied in some detail in Chapter 1. Therefore, a considerably more detailed analysis of cryptomafic volumes can be accomplished for the areas in this region than for any of the other locations listed in Table 1.
       In Chapter 1, a combined volume of ~5 x 105 km3 was obtained for the cryptomafic deposits on the western limb of the Moon, corresponding to 5% of the estimate volume of known mare deposits [Head, 1975b]. Unfortunately, this estimate was hampered by several factors, most importantly, a lack of understanding of the dark halo-forming process and data sets of low resolution. These limitations were subsequently addressed in Chapters 2 and 3 and so a re-evaluation of cryptomafic volume estimates in these areas is now possible. Using the data from Chapter 1 and the results of Chapters 2 and 3, thicknesses of the cryptomafic deposits in these three areas can be re-examined to give improved estimates of thickness and therefore volume.
       It should be noted that these improved thickness estimates are much more detailed than those that can be obtained by comparison to nearby mare deposits. Also, the area estimates for cryptomafic deposits on the western limb, that were determined in the detailed study of Chapter 1, are considerably more accurate than could be obtained by the rough techniques that were outlined in the section above. Therefore, volume estimates for cryptomafic deposits on the western limb are expected to be significantly superior to those from other locations, where comparatively little previous work has been done. However, these variations in the precision and accuracy of cryptomafic volume estimates, for the different cryptomafic regions, are not expected to affect this assessment of the global importance of cryptomafic deposits.

        Schiller-Schickard
       In Chapter 1, a single cryptomafic deposit, filling an old, degraded basin, was located in the Schiller-Schickard region (Chapter 1, Figure 7). Detailed studies of the Schickard crater area in Chapter 3 showed that, in this area, the cryptomafic deposit may be more complex. This deposit is re-evaluated here, using the data from Chapter 1 and results from Chapters 2 and 3
       Following the techniques outlined in Chapter 1, the thickness of the top obscuring layer (tm) is estimated using the smallest observed dark-haloed craters in each 100-km distance interval from the center of Orientale basin (Table 2). The value of tm, calculated using rim-crest diameter measurements (Dr) of Chapter 1 and the equations of Chapter 2, varies from 145 - 317 meters, with an average of roughly 200 meters. In Chapter 3, the thickness of the overlying layer in the Schickard crater area was estimated to be less than 85 meters, suggesting that the data set of Chapter 1 does not allow the smallest dark-haloed craters to be identified. A thickness of 85 meters is, therefore, adopted for the entire obscuring layer in the Schiller-Schickard region.
       Results from Chapter 1 suggest that the largest craters in this area may not be reaching the base of the cryptomafic materials. Depth to the bottom can, therefore, be conservatively estimate from the excavation depths (de) of the largest dark-haloed craters (Table 2). The thickness of cryptomafic materials is estimated by subtracting the thickness of the ejecta layer (85 m) from the calculated de values (Table 2). These estimates range from approximately 600 - 1900 meters, with an average of roughly 1.1 km. In Chapter 3, the cryptomafic deposit in the Schickard area was found to be very thin (~400 m) and a deeper cryptomafic layer was speculated to exist below more than 400 meters of highland material. The presence two cryptomafic layers separated by an intermediate highland layer, was not identified in the data of Chapter 1. Neither was the low thickness in the Schickard area recognized. This may indicate that the cryptomafic layer is highly variable, and that the low number of craters and the bin arrangement chosen (i.e., a function of distance from Orientale as opposed to some geologic criteria) were inadequate to display this variability. In order to insure a conservative estimate, the thickness of 400 meters, identified for the cryptomafic deposit in Chapter 3 is adopted here for the entire Schiller-Schickard area.
       Recalling that the total area of the Schiller-Schickard region was estimated to be 3.8 x 105 km2 in Chapter 1, a conservative estimate of 1.5 x 105 km3 is obtained for the volume of the Schiller-Schickard cryptomafic deposit.
       Some caveats to the volume obtained above must be considered. First, areal estimates of the Schiller-Schickard cryptomafic deposit in Chapter 1 do not exclude areas such as the rims of Schiller and Schickard craters, which are known, or are suspected, to contain no cryptomafic materials [Head et al., 1993b; Hawke and Blewett, 1994]. Second, the Schiller-Zucchius plains, which are thought to represent a shallow mare layer that is underlain by additional cryptomafic layers [Blewett et al., 1995b; Greeley et al., 1993b], were not included in the volume calculations. It is estimated that the erroneous omissions roughly balance out the erroneous inclusions. Also, the thickness value used to estimate the volume was extremely conservative, therefore, the volume calculations noted above should yield a minimum estimate of the volume of cryptomafic materials in this region.
       The ages of the light plains units, which overly the cryptomafic deposit in this region, have been dated as approximately 3.8. b.y. old [Greeley et al., 1993b]. Also, the cryptomafic units is seen to occupy the Schickard crater, which is late pre-Nectarian in age or >3.92 b.y. old [Wilhelms, 1987]. Thus, the cryptomafic layer can be dated at roughly 3.8-3.92 b.y. old.

        Humorum
       The studies of Chapter 1 identified the presence of several cryptomafic deposits in the area west of Mare Humorum (Chapter 1, Figure 10). A large contiguous cryptomafic unit is found just west and concentric around Mare Humorum. Further west, several small cryptomafic patches are also identified.
       One dark-haloed crater in the Humorum region, Gassendi G, has been shown to exhibit the spectral signature of highland materials on the crater floor [Hawke et al., 1993]. This 7.8 km diameter dark-haloed crater has been interpreted to be tapping the bottom of the cryptomafic deposit in this region. Thus, in order to re-assess the thickness of cryptomafic material in the Humorum area, the data from Chapter 1 are re-evaluated using the equations developed in Chapter 2. The thickness of the top-most obscuring layer (tm) is estimated from the smallest dark-haloed craters in each distance interval, and the base of the cryptomafic layer (th) is estimated from the largest dark-haloed craters (Table 3). Estimates of the thickness of the overlying layer range from 120 to 600 meters, with an average of 240 meters. However, it was shown in Chapter 3 that the data set of Chapter 1 does not allow the smallest dark-haloed craters to be identified. To minimize the affects of this limitation, the smallest tm value is assumed to estimate the thickness of the overlying layer for the entire region. The thickness of cryptomafic material is estimated by subtracting this minimum tm from the th values (Table 3), yielding an average thickness of roughly 450 meters. Again, these estimate tend to provide a minimum value since the thickness of the overlying layer may be thinner than estimated by the value of dm.
       Assuming that both the contiguous cryptomafic deposits and the smaller cryptomafic patches that were identified in the area west of Mare Humorum (Chapter 1, Figure 10) can be represented by the minimum thickness estimated above, a conservative estimate of the volume of cryptomafic materials can be calculated. Recalling that the estimated surface area, for all cryptomafic deposits in this region, was found to be 2.3 x 105 km2 (Chapter 1), an minimum estimate of 1.0 x 105 km3 is found for the total volume of cryptomafic material in the Humorum area.
       Orientale ejecta material is extensively mapped in the region [i.e., Wilhelms, 1987], suggesting that the layer overlying the Humorum cryptomafic deposit may contain Orientale ejecta. Thicknesses estimates for the overlying layer (Table 3) are roughly within the range expected for Orientale ejecta at this distance [McGetchin et al., 1973; Schultz et al., 1981], supporting this interpretation. Thus, the age of Orientale basin, 3.8 b.y. [Wilhelms, 1987], is adopted as the upper age boundary of the cryptomafic layer. Considering that some of these cryptomafic units flood the Humorum basin, which is late Nectarian in age and thus younger than 3.92 b.y. old [Wilhelms, 1987], this age is adopted as the lower age boundary of the cryptomafic layer. Cryptomafic materials in the Humorum area range in age from >3.8 to <3.9 b.y.

        Procellarum
       Very few dark-haloed craters were recognized in the area west of Oceanus Procellarum, during the study in Chapter 1. Therefore, the results from this area were not as accurate as in the other regions on the western limb. However, the presence of several small, isolated cryptomafic deposits was identified in the Procellarum area (Chapter 1, Figure 14).
       The paucity of dark-haloed craters in this region makes the practice of binning the craters by distance interval, that was used successfully in the other two regions on the western limb, impractical here. Instead, the overlying layer thickness (tm) and the excavation depths (de) are calculated for each crater, using equations from Chapter 2, and listed in Table 4. As was done in Chapter 1, the minimum value of tm (160 m) was adopted as the thickness of the overlying layer throughout the region and subtracted from the excavation depth of each crater (Table 4), since it is not clear if the bottom of these deposits is being reached. The thickness values of 697, 1008, 982, 1900, and 420 (Table 4) correspond to the maximum thickness values for each of the five cryptomafic patches in this area. The average of these values yields a minimum thickness estimate of 1000 meters of cryptomafic materials in the Procellarum area. Recalling that the total surface area of the Procellarum cryptomafic patches was estimated to be 3.8 x 104 km2 in Chapter 1, the total volume of cryptomafic materials can be estimated to be 3.8 x 104 km3.
       Orientale ejecta is prominently mapped in this region [McCauley, 1967; Scott et al., 1977], and it has been suggested in Chapter 1 that Orientale ejecta is contained in the overlying layer. The upper age bound of the cryptomafic patches in this area is, therefore, 3.8 b.y., the age of the Orientale impact event [Wilhelms, 1987]. The lower age bound for these cryptomafic patches is obtained from the age of Hevelius crater, which contains parts of the largest cryptomafic patch and is adopted as the lower age boundary for all Procellarum cryptomafic patches. The crater Hevelius is identified as Nectarian in age, and thus <3.92 b.y. old [Wilhelms, 1987]. The age of the cryptomafic patches in this area can, therefore, be roughly dated as 3.8 - 3.92 b.y. old.


        Australe
       The Australe basin is a highly degraded, 900 km diameter basin, containing many superposed craters that demonstrate the age of the interior materials [Whitford-Stark, 1979]. Many mare patches are found in and around Australe. These mare patches have a total area of ~ 3.2 x 105 km2, representing approximately 56% of basin surface [Whitford-Stark, 1979]. They are mostly confined to crater floors, but some flooding does occur in intercrater areas, particularly in the northeast, indicating that volcanism here may be extensive.
       Several dark-haloed craters are found in and around Australe basin [Schultz and Spudis, 1979], suggesting the presence of cryptomafic deposits [Whitford-Stark, 1979]. Whitford-Stark [1979] has proposed that mare materials, having a total thickness of 500m - 3.5 km, are located throughout the entire Australe basin. This interpretation is supported by the presence of albedo anomalies (Figure 6), which affect all of Australe basin and extend as far as the outer basin ring [Wilhelms and El-Baz, 1977; Wilhelms et al., 1979]. Topography (Figure 4a) shows the presence of an annular depression, just inside the outer ring area, and the crust is also seen to be thinned (Figure 4b) in approximately this area. Elevated FeO concentrations (Figure 5) are generally found in this region, suggesting the presence of mafic materials throughout the basin. Lunar maps [Wilhelms and El-Baz, 1977; Wilhelms et al., 1979] show the presence of many light plains materials, of various ages, occupying the basin floor. Furthermore, the presence of younger basalts towards the basin edges suggests that some extension is occurring from loading near the basin center, and the presence of mare ridges suggests downwarping in the basin center, also from mare loading [Whitford-Stark, 1979]. All of the above evidence strongly argues for the existence of an extensive cryptomafic deposit in this area, filling the whole of the basin floor. Whitford-Stark [1979] has suggested that mafic thicknesses at the locations of the visible maria exceed the measured thicknesses of surface mare units. Therefore, cryptomafic deposits are expected to underlie not only the light plains units, but also the visible mare patches as well.
       On the basis of these observations, it is suggested that the area of the cryptomafic unit in Australe can be estimated by considering the area of a 900 km diameter circle, representing the Australe basin. In this manner, the area of the Australe cryptomafic deposit is calculated to be 6.4 x 105 km2. Whitford-Stark [1979] has estimated that the thickness of mare materials in this area ranges from 0.5 - 3.5 km. A conservative estimate of 500 meters is, therefore, adopted for the thickness of the cryptomafic unit here. Multiplying this thickness estimate by the areal estimate given above yields a volume estimate of 3.2 x 105 km3 for the Australe cryptomafic deposit.
       Lunar maps of this area [Wilhelms and El-Baz, 1977; Wilhelms et al., 1979] show that the plains material on the floor of Australe basin has a wide range of ages. Many light plains deposits are mapped as Nectaris ejecta. This suggests that some of the cryptomafic materials in Australe may be as old as pre-Nectarian in age, or >3.92 b.y. [Wilhelms, 1987]. However, not all of the cryptomafic deposits here are expected to be this ancient. The various ages of the visible mare patches in this area illustrate the extended nature of extrusive volcanism in this region [Whitford-Stark, 1979]. Therefore, it is expected that the range of cryptomafic ages will also vary, and so the presence of more than one cryptomafic layer is inferred. The youngest cryptomafic deposits in this region are expected to be pre-Orientale in age, or >3.8 b.y. old [Wilhelms, 1987] , since the youngest lights plains units here are identified as Imbrium or Orientale ejecta [Wilhelms and El-Baz, 1977; Wilhelms et al., 1979]. The lower bound of these younger cryptomafic deposits is expected to be the Nectaris ejecta deposit (3.92 b.y. old [Wilhelms, 1987]), which is suspected to be relatively thick here [Whitford-Stark, 1979]. The lower bound for the older, pre-Nectarian cryptomafic units can be represented by the Australe basin, which these units occupy. Australe basin is identified as early pre-Nectarian, and so can be dated as roughly <4.0 b.y. old [Wilhelms, 1987]. In summary, at least two cryptomafic layers are inferred in the Australe basin area, with ages ranging from 3.92 - 4.0 b.y. for the older unit, and 3.8 - 3.92 b.y. old for the younger unit. The relative volumes of these two cryptomafic units are not known.


        Balmer
       The Balmer cryptomafic area occupies a region known as the Balmer basin, a degraded, pre-Nectarian basin, with an inner ring measuring 225 km in diameter and an outer ring 450 km in diameter [Maxwell and Andre, 1981]. One small mare patch is located to the west of this area and several large, young craters have emplaced ejecta deposits within the Balmer basin [Wilhelms and El-Baz, 1977].
       Many lines of evidence indicate the presence of cryptomafic deposits in this region. Dark-haloed craters have been found here, most notably on the ejecta deposits of Langrenus and Petavius craters, indicating the presence of underlying mafic materials [Schultz and Spudis, 1979, 1983; Hawke and Spudis, 1980; Antonenko et al., 1995]. Spectral analysis has further confirmed that basalt excavation is occurring northeast of Petavius crater [Bell and Hawke, 1984]. Geochemical analysis has shown that elevated Th concentrations and Mg/Al ratios [Frontispiece, 1977; Maxwell and Andre, 1981; Clark and Hawke, 1987], as well as higher FeO and TiO2 values with respect to surrounding areas [Giguere et al., 1998], can be found at this location, suggesting the presence of hidden mafic materials. Furthermore, the presence of a low albedo zone in the Balmer area has been recognized by Greeley et al. [1993a]. All of this evidence suggests the presence of buried mafic materials and a cryptomafic deposit, obscured by crater dusting, has been proposed for this area [Hawke et al., 1985; Antonenko et al., 1995].
       The extent of this cryptomafic deposit can be estimated by considering the evidence in more detail. Th and Mg/Al anomalies in the Balmer basin are generally associated with light plains units [Hawke at al., 1985; Clark and Hawke, 1987], corresponding to the INp and some Ip plains units [Wilhelms and El-Baz, 1977; Hawke et al., 1985]. Elevated FeO values (Figure 5) generally describe a roughly circular, irregular area. An irregular shape for this cryptomafic deposit is consistent with the topography and crustal thickness (Figure 4) in this area, which is characterized by irregularly shaped topographic lows and variations in crustal thickness. The albedo anomaly (Figure 6) is also irregular in shape, located roughly between the craters Petavius, Langrenus, and Ansgarius. However, the cryptomafic deposit is know to extend beneath the ejecta deposits of Petavius and Langrenus, because of the presence of dark-haloed craters here [Schultz and Spudis, 1979, 1983; Hawke and Spudis, 1980; Antonenko et al., 1995]. On the basis of these observations, the area of the Balmer cryptomafic deposit is estimated as roughly 1.7 x 105 km2. From their studies of the eastern maria, De Hon and Waskom [1976] concluded that maria on the eastern limb tend to be 200-400 meters thick on average, but may be as thick as 1200-1500 meters in basin centers. A conservative estimate of 400 meters is, therefore, adopted for the thickness of the entire Balmer cryptomafic unit. Multiplying this thickness by the areal estimate yields a volume of 6.8 x 104 km3 for the cryptomafic deposit in the Balmer basin.
       Geochemical anomalies in the Balmer basin have been associated with INp and Ip light plains units [Hawke at al., 1985; Clark and Hawke, 1987]. These units are identified as late Nectarian or early Imbrian in age [Wilhelms and El-Baz, 1977], or roughly >3.87 b.y. old [Wilhelms, 1987]. The age of these units is, therefore, adopted as the upper age bound for the cryptomafic deposit at this location. The lower age bound can be represented by the Balmer basin, which the cryptomafic deposit occupies. The Balmer basin is an early pre-Nectarian basin, dated at roughly <4.1 b.y. old [Wilhelms, 1987]. Cryptomafic materials in the Balmer basin region can thus be dated as roughly 3.87 - 4.1 b.y. old.


        Frigoris
       A multitude of light plains units, of various ages, have been mapped in the areas north of Mare Frigoris [Lucchitta, 1978]. However, not all of these plains deposits are expected to obscure mafic materials. Spectral studies of the far northern light plains, for example, appear to indicate only highland materials at these locations [Belton et al., 1994], or else produce ambiguous results [Mustard and Head, 1995]. In the areas northeast of Mare Frigoris, however, the presence of cryptomafic units has been proposed. The Imbrian aged light plains here have been interpreted to consist of thin Orientale ejecta, which overlies mare deposits [Lucchitta, 1978], indicating the presence of cryptomafic material at this location [Whitford-Stark, 1990]. Using spectral mixing analysis of Galileo data, some of these light plains have been shown to contain a significant mare component, namely areas at the eastern boundary of Mare Frigoris and between Mare Frigoris and Kane crater [Mustard and Head, 1995].
       While the dark-haloed crater map of Schultz and Spudis [1979] shows no dark-haloed craters north-east of Mare Frigoris (Figure 2), subsequent studies have identified dark-haloed craters in the light plains of this region [Giguere et al., 1998]. The crater Gärtner D (8 km diameter) has been identified as a dark-haloed crater, and the excavation of basalts by this crater has been confirmed spectrally [Belton et al., 1994; Greeley et al., 1993a; Mustard and Head, 1995]. However, the plains immediately adjacent to Gärtner D have highland spectra [Mustard and Head, 1995], suggesting that the overlying ejecta deposits may be locally thick. A few immature dark-haloed craters have also been identified spectrally in the Frigoris area [Head et al., 1993a].
       Information on the distribution of dark-haloed craters and light plains units that contain mafic materials can be used to define the extent of the Frigoris cryptomafic. Other data can also be used to corroborate these determinations. A low albedo zone north of eastern Frigoris has been previously identified [Greeley et al., 1993a] and can be confirmed in the albedo map of Figure 6. This albedo low is generally associated with Ip2 units [Lucchitta, 1978]. Enhanced FeO values from Clementine data have also been previously identified [Giguere et al., 1998]. However, the FeO map of Figure 5, shows that the FeO anomaly does not extend very far beyond the boundaries of Mare Frigoris. The presence of patchy crustal thickness depressions and an extended topographic low in this area can be seen in Figure 4.
       On the basis of these observations, the area of the Frigoris cryptomafic deposit is estimated to be roughly 3.7 x 105 km2. Whitford-Stark [1990] has estimated that the thickness of mare materials in Mare Frigoris is >400 meters. The value of 400 meters is, therefore, adopted for the thickness of cryptomafic materials in the area northeast of Mare Frigoris. Multiplying this thickness by the areal estimate given above yields a volume estimate of 1.5 x 105 km3 for the Frigoris cryptomafic deposit.
       The light plains units in this region have been identified as originating from the Orientale basin [Lucchitta, 1978], thus the age of Orientale, 3.8 b.y. [Wilhelms, 1987], can be used as an upper age bound for the cryptomafic deposits of this region. Furthermore, the Frigoris cryptomafic deposit occupies several pre-Nectarian aged craters, such as GŠrtner and Kane, whose rims protrude above the light plains units. Thus, a pre-Nectarian age of >3.92 b.y. [Wilhelms, 1987] can be defined as the lower age boundary for the cryptomafic deposit in this area. The Frigoris cryptomafic deposit can, therefore, be roughly dated at 3.8-3.92 b.y. old.


        Langemak
       Langemak crater (8º S, 118º E) is a young, Imbrian aged crater located on the southeastern rim of the Al-Khwarizmi-King basin [Wilhelms and El-Baz, 1977]. Al-Khwarizmi-King is a very old, degraded, pre-Nectarian basin, approximately 590 km in diameter, which is located on the eastern limb of the Moon [Wilhelms, 1987]. This basin contains many old, degraded craters, and its floor is covered with rugged plains materials [Wilhelms and El-Baz, 1977].
       Several dark-haloed craters have been located in the vicinity of Langemak crater, and Langemak itself has been identified as a dark-haloed crater [Schultz and Spudis, 1979]. Geochemical anomalies, indicating the presence of mafic materials, have been identified in Langemak ejecta [Hawke et al., 1985]. However, the anomalies are not limited to the immediate vicinity of Langemak crater [Hawke and Spudis, 1980]. High Fe and Ti values have been identified in much of the Al-Khwarizmi-King basin area [Hawke and Spudis, 1980]. High Mg values [Clark and Hawke, 1991] have been associated with areas mapped as "bright sinuous markings" [Wilhelms and El-Baz, 1977] in the northern portions of the basin. Mafic signatures were also found in light plains units near Langemak [Clark and Hawke, 1991]. In Pasteur crater, geochemical anomalies correlate to the INp plains units [Wilhelms and El-Baz, 1977; Hawke et al., 1985]. All of this evidence strongly suggests the presence of extensive cryptomafic deposits in this area. Several authors have even proposed that basalts underlie most of the Imbrian and Nectarian light plains in the Al-Khwarizmi-King basin [Hawke, 1979; Hawke and Spudis, 1980], as well as some unmapped light plains deposits [Hawke et al., 1985].
       On the basis of these observations, the cryptomafic deposit in the Langemak area is predicted to occupy the entire Al-Khwarizmi-King basin and to extend south into the regions in and around Langemak and Pasteur craters. Other data did not prove useful in corroborating this prediction. No FeO anomaly could be seen in the data of Figure 5. Only a slight topographical depression and decrease in crustal thickness were observed in Figure 4. The albedo anomaly in this region (Figure 6) was very minor, almost non-existent, and so did not help in delimiting the extent of the cryptomafic deposit. Therefore, the dimensions of the cryptomafic deposit that were determined from the literature are adopted here. The area of the cryptomafic deposit, so determined, is estimated to be roughly 4.0 x 105 km2. De Hon and Waskom [1976] calculated that thicknesses of the eastern maria are 200-400 meters on average, and may be as thick as 1200-1500 meters in basin centers. A conservative estimate of 400 meters is, therefore, adopted for the thickness of the of the entire cryptomafic deposit in the Langemak area. When multiplied by the areal estimate determined above, this thickness suggests that the Langemak cryptomafic deposit represents a volume of 1.6 x 105 km3.
       Most of the plains deposits occupying the floor of Al-Khwarizmi-King basin have been dated as pre-Nectarian in age [Wilhelms and El-Baz, 1977]. Since these plains overly the cryptomafic deposit in this area, their age (<3.92 b.y. old [Wilhelms, 1987]) can be used as an upper age bound for the cryptomafic deposit. This interpretation is consistent with previous suggestions that a pre-Nectarian mafic deposit is located in the Al-Khwarizmi-King basin [Spudis and Davis, 1983]. For the lower age bound, the age of the Al-Khwarizmi-King basin, which the cryptomafic units occupy, can be adopted. The Al-Khwarizmi-King basin has been dated as very early pre-Nectarian, roughly 4.1 b.y. old [Wilhelms, 1987]. Therefore, the age of the Langemak cryptomafic deposit can be estimated as 3.92 - 4.1 b.y.


        Smythii
       The Smythii basin is an old, 840 km diameter, pre-Nectarian basin, located on the eastern limb of the Moon [Wilhelms, 1987]. Several small mare patches are dispersed in and around the inner ring (360 km in diameter) of the basin, with one large mare deposit occupying the central northeast portion [Wilhelms and El-Baz, 1977]. A large furrowed and pitted plains unit is located inside the inner ring and many small light plains patches are scattered outside the inner ring [Wilhelms and El-Baz, 1977].
       A large number of dark haloed craters are located all around the Mare Smythii area (Figure 2). Specifically, dark-haloed craters are found in the plains units of the Smythii basin floor [Schultz and Spudis, 1979], where they excavate basalts from a depth of several hundred meters [Conca and Hubbard, 1979]. Geochemical evidence also suggests that mafic material has been mixed into the plains deposits [Andre et al., 1979]. Mafic signatures have been found in plains units in the center of Smythii basin, as well as in craters such as Babcock that are located outside the inner Smythii ring [Clark and Hawke, 1991; Hawke and Spudis, 1980]. The existence of an early volcanic unit in Smythii basin, which underlies the visible mare and plains deposits alike, has been proposed for this region [Conca and Hubbard, 1979].
       The previous studies, as discussed above, suggest that cryptomafic deposits in Smythii basin are located mainly within the inner basin ring, with a scattering of smaller cryptomafic patches in the area between the inner and outer rings. This interpretation is confirmed by the albedo map of Figure 6 and the FeO map of Figure 5, which illustrate the presence of large anomalies inside the inner ring, but only minor or no anomalies outside the inner basin ring. This is also consistent with topography and crustal thickness data (Figure 4), which show that crustal thinning and topography lows do no extend much beyond the inner ring of Smythii basin.
       Using the light plains units mapped by Wilhelms and El-Baz [1977] as a guide, the total area of the Smythii cryptomafic deposits is estimated to be 2.0 x 105 km2. On the basis of Conca and Hubbard [1979] the existence of unrecognized mafic materials is assumed and these are included in the areal estimate. In the analysis of previous cryptomafic areas on the eastern limb, the value of 400 meters was adopted for the thickness of cryptomafic units on the basis of studies conducted by De Hon and Waskom [1976]. This value is, therefore, adopted for the thickness of the Smythii cryptomafic deposits as well. Multiplying this thickness by the areal estimate determined above yields a combined volume estimate of 8.0 x 104 km3 for all cryptomafic patches in the Smythii basin.
       The age of the light plains units that define the Smythii cryptomafic deposits has been dated as Imbrian aged [Wilhelms and El-Baz, 1977]. Furthermore, ejecta from a range of Imbrian aged craters, such as Kiess, Widmannstatten, Helmet, and Warner, has been identified as overlying this region [Conca and Hubbard, 1979]. The age of approximately 3.8 b.y., the boundary between the upper and lower Imbrian series, is therefore adopted as the upper age bound for the cryptomafic deposits in this region. The lower age bound for the Smythii cryptomafic deposits can be obtained by considering the rugged plains material between the inner and outer rings of Smythii. These plains are dated as Nectarian or pre-Nectarian in age [Wilhelms and El-Baz, 1977]. Smythii basin is pre-Nectarian in age, but the nearby Crisium basin is dated as Nectarian and so younger than Smythii basin [Wilhelms, 1987]. The ejecta deposits of Crisium basin are, therefore, expected to blanket the Smythii region and the rugged plains that occupy the area between the inner and outer rings of Smythii basin may represent this ejecta material. For this reason, the age of Crisium basin, which is estimated to be 3.84 b.y. [Wilhelms, 1987], is taken as the lower age boundary for the cryptomafic deposits in this region. While it is possible that the Crisium ejecta deposit may obscure even earlier mafic materials, no evidence has been found for such units. Evidence for cryptomafic deposits in this area is generally confined to the younger plains units [Clark and Hawke, 1991; Hawke and Spudis, 1980]. Cryptomafic deposits here must, therefore, post-date the Crisium impact event and the age of Smythii cryptomafic deposits can be given as roughly 3.8-3.84 b.y. old.


        Taruntius
       Taruntius crater is located in a highland region that borders the contact between Mare Tranquilitatis and Mare Fecunditatis. Taruntius has been dated as a Copernican aged crater, but it is surrounded by light plains materials that are dated as Imbrian in age [Wilhelms and McCauley, 1971].
       A dark-haloed crater has been identified on the rim of Taruntius crater [Schultz and Spudis, 1979; Hawke et al., 1985; Giguere et al., 1998] suggesting the presence of cryptomafic deposits in this area. Spectral mixing analysis of telescopic data [Blewett et al., 1995a] and mixing calculations [Hawke and Spudis, 1980; Hawke et al., 1985] have demonstrated that mare materials are present in the Taruntius area, and that they are generally associated with light plains deposits (It and Ip units [Wilhelms and McCauley, 1971]). The morphology of the plains units is consistent with volcanism [Hawke, 1979]. This interpretation is further confirmed by the existence of geochemical anomalies [Frontispiece, 1978] and albedo anomalies [Pohn and Wildey, 1970] in these areas. Clementine data has also shown the presence of high FeO and TiO2 values in the light plains areas around Taruntius crater [Giguere et al., 1998]. It has been suggested that Taruntius crater excavated highland material from beneath Mare Tranquillitatis and Mare Fecunditatis, obscuring these maria with high-albedo, highland materials [Giguere et al., 1998; Hawke et al., 1985].
       The cryptomafic deposit in the Taruntius area appears to be related to the various plains units in this region (It and Ip units [Wilhelms and McCauley, 1971]) and with the ejecta of Taruntius crater. Albedo anomalies (Figure 6) appear blotchy, corresponding to light plains units. FeO values (Figure 5) are generally high in the region between Mare Tranquillitatis and Mare Fecunditatis, showing no indication of a low Fe highland area between the two maria. Topography and crustal thickness data (Figure 4) also show generally low values, more consistent with the surrounding maria than nearby highland areas. On the basis of these observations, the area of the Taruntius cryptomafic deposit is estimated from the area of the light plains deposits in this region. So estimated, the cryptomafic deposit around Taruntius crater is found to be 8.7 x 104 km2.
       The thickness of Mare Tranquilitatis has been estimated by several authors. De Hon [1974] calculated that Mare Tranquillitatis is, on average 500-600 meters thick. However, using crater penetration techniques, Staid et al. [1996] estimate the thickness in Tranquillitatis to be ~2 km. They further demonstrated that even at the edges of the mare, thicknesses of ~600 meters are common [Staid et al., 1996]. Therefore, the value of 600 meters is adopted for the thickness of the cryptomafic deposit in the Taruntius region, located just south of Mare Tranquillitatis. Multiplying this thickness by the areal estimate obtained above, the volume of cryptomafic material in the Taruntius area is calculated to be 5.2 x 104 km3.
       The age of the light plains that define the Taruntius cryptomafic deposit have been dated as Imbrian in age [Wilhelms and McCauley, 1971]. The adjacent Maria Tranquillitatis and Fecunditatis are also Imbrian in age [Wilhelms, 1987]. The hypothesis of Giguere et al. [1998] that the cryptomafic deposit in this region represents obscured Mare Tranquillitatis and Mare Fecunditatis materials is adopted here. The oldest age (~3.8 b.y. old [Wilhelms, 1987]) of the Apollo 11 samples, which were returned from Mare Tranquillitatis, is therefore assumed for the Taruntius cryptomafic deposit, since the maria immediately adjacent to Taruntius crater are stratigraphically some of the oldest in the region [Wilhelms, 1987]. The age of the Taruntius cryptomafic deposit is thus roughly dated at 3.8 b.y. old.


        South Pole-Aitken
        The South Pole-Aitken basin is the largest identified basin on the Moon, located on the south-central farside and measuring approximately 2500 km in diameter [Wilhelms, 1987]. Clementine data shows that the lowest elevations on the Moon are found in this basin [Spudis et al., 1994], as well as some of the thinnest crust [Zuber et at., 1994; Neumann et al., 1996]. Many large and degraded craters are superposed on top of the South Pole-Aitken basin [Wilhelms, 1987], attesting to its extreme age. Also, many small mare patches are found in South Pole-Aitken [Wilhelms, 1987], representing much of the far side mare deposits.
       Several different lines of evidence suggest that a mafic anomaly exists in the South Pole-Aitken basin. Dark-haloed craters have been identified in the northern areas of the basin [Schultz and Spudis, 1979, 1983]. An enriched iron content was recognized from the Apollo data [Metzger et al., 1974; Stuart-Alexander, 1978], and more recently confirmed from the Clementine data [Lucey et al., 1995, 1998a; Blewett et al., 1997]. A magnetic anomaly has been distinguished in the South Pole-Aitken basin [Hawke and Spudis, 1980]. The albedo in the basin is anomalously low [Belton et al., 1992; Head et al., 1993b] And analysis of Galileo UVVIS filter ratios shows the presence of an enhanced mafic component in South Pole-Aitken [Belton et al., 1992].
       However, the existence of cryptomafic deposits in the South Pole-Aitken basin is not commonly accepted. Some authors have suggested that thick cryptomafic deposits have been obscured by ejecta from subsequent impacts, such as Schrödinger, Australe, and Apollo, in this region [Head et al., 1993b]. Others invoke different explanations for the anomalies observed. Several authors suggest that this large basin is excavating the lower crust or upper mantle and that these more mafic deep strata can account for the observed mafic signatures [Belton et al., 1992; Greeley et al., 1993b; Head et al., 1993b; Pieters et al., 1993; Lucey et al., 1998b]. McEwen et al. [1993] claim that the basin floor exhibits spectra that are similar to impact melts, although the an interpretation based on the mixing of mare and highland materials cannot be ruled out [Lucey et al., 1998b]. Pieters et al. [1997] argue that the spectral signatures in South Pole-Aitken are not consitant with either impact melt, cryptomare, or mantle materials but rather are more consistent with a mafic lower crust. Thus, the character of the mafic anomaly on the floor of South Pole-Aitken is controversial.
       Areas in the northern parts of South Pole-Aitken are associated with dark-haloed craters [Schultz and Spudis, 1979, 1983] and geochemical anomalies [Metzger et al., 1974; Stuart-Alexander, 1978]. Thus, the presence of cryptomafic deposits has been suggested in the region north of Mare Ingenii and near Van de Graaff crater [Hawke and Spudis, 1980]. These deposits correspond to an area of approximately 2.5 x 105 km2.
       Maria on the eastern limb of the Moon tend to have an average thickness of 200-400 meters [De Hon and Waskom, 1976]. Western maria tend to be ~ 400 meters thick on average [De Hon, 1979]. Mare materials in nearby Mare Australe have been estimated to range in thickness from 500 meters - 3.5 km [Whitford-Stark, 1979]. On the basis of these considerations, a conservative thickness estimate of 400 meters is adopted for the cryptomafic deposit in South Pole-Aitken basin. Multiplying this by the areal estimate determined above yields a conservative volume estimate of 1 x 105 km3.
       The ages of mare deposits in Van de Graaff and Apollo basins has been dated at 3.64 b.y. and 3.63 b.y., respectively [Greeley et al., 1993b]. Since these mare deposits have not been obscured, it can be assumed that cryptomafic units that have been obscured are older than these ages. The South Pole-Aitken basin, which contains the cryptomafic units, can be considered for the lower age boundary. This basin has been dated at approximately 4.1 b.y. [Wilhelms, 1987]. A rough age estimate of the proposed South Pole-Aitken cryptomafic deposit is, therefore, 3.63 - 4.1 b.y. The wide range in ages for the plains deposits that occupy the floor of South Pole-Aitken [Stuart-Alexander, 1978; Wilhelms et al., 1979] suggests more than one episode of cryptomafic volcanism may have occurred here, as was the case in nearby Australe basin. The bulk of the plains materials in South Pole-Aitken are dated as Nectarian or pre-Nectarian in age [Stuart-Alexander, 1978; Wilhelms et al., 1979], roughly 3.9 b.y. old [Wilhelms, 1987]. I, therefore, propose that two cryptomafic layers may be located in the northern parts of South Pole-Aitken basin, one ancient layer dated at roughly 3.9 - 4.1 b.y. old and a younger layer dated at roughly 3.63 - 3.9 b.y. old. The greater surface area of the older plains units [Stuart-Alexander, 1978; Wilhelms et al., 1979] suggests that the older cryptomafic deposits will be volumetrically more significant.


        Schrödinger
       The Schrödinger basin is a 320 km diameter, early Imbrian aged basin. Schrödinger is located within the South Pole-Aitken basin, but it does not occur within the albedo anomaly that defined the South Pole-Aitken cryptomafic deposits and so was not considered in the analysis of that deposit. Therefore, Schrödinger basin is briefly discussed here.
       The data returned by the Clementine mission made it possible for Schrödinger basin to be mapped in detail for the first time [Shoemaker et al., 1994]. Smooth plains and ridged plains (sp and rp units) were mapped and interpreted to be melt sheets. However, the presence of ghost craters and one lobate ridge were identified in these units [Shoemaker et al., 1994], suggesting a mare-like morphology. Post-sp unit mare patches were also identified [Shoemaker et al., 1994] attesting to the possibility of extrusive volcanism in this area. The presence of volcanic dark-haloed craters, which often occur around edges of basin filling maria [Coombs and Hawke, 1992], was noted [Shoemaker et al., 1994]. Radial graben, which are generally associated with basin subsidence due to mare loading [Solomon and Head, 1979, 1980] were also identified [Shoemaker et al., 1994]. All of this evidence seems to argue for the presence of a small cryptomafic unit in the center of Schrödinger basin.
       A weak albedo anomaly (Figure 6) was identified in the Schrödinger area, corresponding to the plains materials mapped by Shoemaker et al. [1994]. A slight topographic and crustal thickness anomaly (Figure 4) can be identified associated with Schrödinger basin, but no FeO anomaly (Figure 5) could be distinguished. Therefore, on the basis of the plains units mapped by Shoemaker et al. [1994] a cryptomafic unit, which is roughly circular with a diameter of ~80 km, is proposed in this region. The area of this cryptomafic deposit is estimated to be 5 x103 km2. A very conservative thickness estimate of 200 meters is assumed on the basis of mare thickness estimates from De Hon and Waskom [1976]. Considering this thickness and areal estimate, a volume of 1 x 103 km3 is determined. This cryptomafic deposit appears to be volumetrically very insignificant.
       The early Imbrian age of the Schrödinger basin places a lower age boundary of approximately 3.87 b.y. on the age of this small cryptomafic deposit. The presence of superposed mare units on top of the cryptomafic deposit places an upper age bound of very roughly 3.1 b.y. old, the youngest age of returned basalt samples from the Moon [Wilhelms, 1987]. The Schrödinger cryptomafic deposit is, therefore, very roughly dated at 3.1 - 3.87 b.y. old.


        Mendel-Rydberg
       The Mendel-Rydberg basin is a Nectarian aged basin that is approximately 630 km in diameter [Wilhelms, 1987]. It is located on the southwestern limb of the Moon, directly south of Orientale basin. Deposits of Orientale ejecta cover much of the Mendel-Rydberg basin, and light plains deposits have been mapped in the south-east quadrant [Wilhelms et al., 1979]. Two mare patches are located in the center of the basin [Wilhelms et al., 1979].
       One dark-haloed crater (Figure 2) was identified in this area by Schultz and Spudis [1979, 1983], but it was not until the Galileo Earth-Moon 2 flyby imaged the western limb of the Moon that the existence of a substantial cryptomafic deposit was suspected here [Head et al., 1993b; Mustard et al., 1992]. Galileo filter ratio analysis identified the presence of an enhanced mafic component in the Mendel-Rydberg basin [Belton et al., 1992]. Clementine data supports the interpretation of a cryptomafic deposit here. Laser altimetry profiles of the Mendel-Rydberg basin have been shown to be consistent with partial filling by lavas, followed by obscuration from Orientale ejecta [Spudis et al., 1994]. A mascon has been identified in Mendel-Rydberg [Neumann et al., 1996] and the crustal thickness and topography maps (Figure 4) illustrate the presence of a clear basin structure. Only a very minor FeO anomaly (Figure 5) is identified in the basin area, but a clear albedo anomaly (Figure 6) is associated with the region inside the inner ring of Mendel-Rydberg [Wilhelms et al., 1979].
       Head et al. [1993b] and Mustard et al. [1992] suggested the existence of an approximately 200 km diameter cryptomafic deposit in the Mendel-Rydberg basin, confined by the basin's inner ring. Observations from this study are consistent with this finding. Therefore, the area of the Mendel-Rydberg cryptomafic deposit is estimated on the basis of a 200 km diameter, roughly circular deposit, yielding a value of 3.1 x 104 km2. Thicknesses of the western maria were evaluated by De Hon [1979], where it was found that thicknesses can range up to 1500 meters in basin centers, but generally average about 400 meters. A conservative estimate of 400 meters is, therefore, adopted for the thickness of the cryptomafic deposit in Mendel-Rydberg basin. Multiplying this thickness by the areal estimate obtained above yields a volume estimate of 1.2 x 104 km3 for the Mendel-Rydberg cryptomafic deposit.
       A pre-Orientale age for the Mendel-Rydberg cryptomafic deposit has been previously proposed by Head et al. [1993b] and Mustard et al. [1992] due to the presence of Orientale ejecta in this region. This interpretation is adopted here, giving an upper age bound of 3.8 b.y., the age of the Orientale impact event [Wilhelms, 1987]. A lower age bound can be obtained from the Mendel-Rydberg basin, which contains the cryptomafic deposit. Mendel-Rydberg is a Nectarian aged basin, dated at approximately 3.92 b.y. [Wilhelms, 1987]. The age of the Mendel-Rydberg cryptomafic deposit can, therefore, be roughly estimated as 3.8 - 3.92 b.y. old.


        Marginis
       Mare Marginis is located on the eastern limb of the Moon, directly north and bordering the Smythii cryptomafic area. In order to prevent any cryptomafic units from being considered twice, a clear, but arbitrary, distinction was made between areas that belong to the Smythii and the Marginis cryptomafic deposits. All areas south of Mare Marginis were assigned to the Smythii region, while all other areas were relegated to the Marginis region.
       Mare Marginis is associated with the pre-Nectarian Marginis basin. However, unlike most basin filling maria, the Marginis Mare is located off-center on the southern edge of the basin. Furrowed and pitted plains material (unit INfp) occupies the basin center [Wilhelms and El-Baz, 1977] suggesting that cryptomafic deposits may be located here. Dark-haloed craters have been found in the region north of Mare Marginis [Schultz and Spudis, 1979], however, most of the cryptomafic evidence is focussed on the light plains units directly east of the mare, where dark-haloed craters have also been identified. High Ti concentrations have been noted in these light plains units [Hawke and Spudis, 1980] and a low albedo zone corresponding to these eastern plains has also been observed [Greeley et al., 1993a]. Figure 6 shows that a moderate low albedo anomaly is also present in the regions north of Mare Marginis, were rays from younger craters have obscured some areas. An FeO anomaly (Figure 5) is present in the Marginis area, extending well beyond the mare boundaries both to the north and south. Topography and crustal thickness anomalies (Figure 4) are observed in the area, but these are irregular and patchy, suggesting an irregular cryptomafic unit.
       On the basis of these observations, two separate cryptomafic units are proposed in the Marginis area. One patch corresponds to the light plains units to the east of Mare Margins. Its area is estimated to be roughly 2.9 x 104 km2. Another patch occupies the region immediately north of Mare Marginis, corresponding to the southern areas of the furrowed and pitted plains [Wilhelms and El-Baz, 1977] The area of this cryptomafic unit is estimated to be 6.7 x 104 km2. No surface mare deposits are included in these areal estimates. As was done for previous cryptomafic deposits on the eastern limb of the Moon, a thickness of 400 meters is adopted, based on the work of De Hon and Waskom [1976]. Multiplying this thickness by the areal estimates obtained above yields a volume of 1.2 x 104 km3 for the cryptomafic deposit east of Mare Marginis and 2.7 x 104 km3 for the cryptomafic deposit north of Mare Marginis.
        The Marginis basin is a very old basin that has been dated at approximately 4.1 b.y. old [Wilhelms, 1987]. The age of this basin provides a lower bound for the age of the cryptomafic deposits that occupy this region. The upper age bound for these cryptomafic units is estimated from the deposits that overly them. The eastern cryptomafic unit in this area is obscured by light plains deposits that have been dated as early Imbrian [Wilhelms and El-Baz, 1977] and therefore roughly >3.8 b.y. [Wilhelms, 1987]. The northern cryptomafic unit is obscured by furrowed and pitted plains that are dated as Nectarian or Imbrian in age [Wilhelms and El-Baz, 1977], or roughly 3.87 b.y. old [Wilhelms, 1987]. Thus, cryptomafic deposits in the Marginis basin can be roughly dated at 3.87 - 4.1 b.y. old for the deposit north of Mare Marginis and 3.8 - 4.1 b.y. old for the deposit east of Mare Marginis.


        Lomonosov-Flemming
       The Lomonosov-Flemming basin is located on the eastern limb of the Moon, to the east of Marginis basin and north of Al-Khwarizmi-King basin. This pre-Nectarian basin is approximately 620 km in diameter [Wilhelms, 1987]. It contains several light plains units of different ages, but no mare deposits [Wilhelms and El-Baz, 1977].
       Very little evidence exists for the identification of cryptomafic deposits in this area. Several dark-haloed craters (Figure 2) have been observed in the Lomonosov-Flemming basin [Schultz and Spudis, 1979], and a cryptomafic deposit has been suggested on the basis of Galileo data [Belton et al., 1994]. Minor topographic and crustal thickness lows can be seen in Figure 4, but these are irregular and patchy. Only minor albedo anomalies (Figure 6) are present in the Lomonosov-Flemming area, and no FeO anomaly can be distinguished (Figure 5). However, on the basis of the dark-haloed crater and Galileo data, the presence of a cryptomafic deposit is proposed here, corresponding to the Ip and Np plains units in the region [Wilhelms and El-Baz, 1977]. On this basis, the area of the cryptomafic deposit in Lomonosov-Flemming is estimated to be 5.2 x 104 km2. Adopting a thickness of 400 meters for this cryptomafic unit, based on average mare thicknesses for the eastern limb [De Hon and Waskom, 1976], the volume of the Lomonosov-Flemming cryptomafic deposit can be estimated as approximately 2.1 x 104 km3.
       The age of the Lomonosov-Flemming cryptomafic deposit can be estimated by considering the ages of the plains units that define the deposit and the basin that contains them. The Lomonosov-Flemming basin is early pre-Nectarian and approximately 4.0 b.y. old [Wilhelms, 1987]. The Ip and Np plains units are dated as lower Imbrian to Nectarian in age [Wilhelms and El-Baz, 1977] and therefore >3.8 b.y. old [Wilhelms, 1987]. The age of the cryptomafic deposit in Lomonosov-Flemming basin is roughly dated as 3.8 - 4.0 b.y old.


        Hercules
       Hercules crater (42º N, 39.7º E) is located at the southeastern tip of Mare Frigoris. The crater itself contains mare materials and is bordered by mare deposits to the west [Scott, 1972a; Grolier, 1974; Lucchitta, 1978]. To the south, extensive light plains units are found, dividing the deposits of Lacus Mortis from Lacus Somniorum [Scott, 1972a; Grolier, 1974].
       Several dark-haloed craters have been identified in the light plains units to the south of Hercules, and one dark-haloed crater is found on the southern portions of the Hercules ejecta deposit [Giguere et al., 1998]. Spectral analysis has demonstrated that basalt excavation is occurring in these craters [Bell and Hawke, 1984], thus arguing for the existence of cryptomafic deposits south of Hercules crater. It has been suggested that ejecta from Hercules and nearby Atlas craters has obscured the surrounding mafic materials, forming a cryptomafic deposit [Giguere et al., 1998]. The presence of cryptomafic materials south of Hercules is confirmed by the albedo map of Figure 6. Albedo anomalies are identified and seen to be associated with Ip, IpIt, and It units [Scott, 1972a; Grolier, 1974]. Based on the extent of these light plains units, an areal estimate of 6.2 x 104 km2 is determined for the cryptomafic deposit south of Hercules. Since the thickness of mare materials in nearby Mare Frigoris has been found to be >400 meters [Whitford-Stark, 1990], a value of 400 meters is adopted for the thickness of cryptomafic materials here. Multiplying this thickness by the areal estimate determined above, the volume of the Hercules cryptomafic deposit is estimated to be 2.5 x 104 km3.
       The age of the light plains that obscure the cryptomafic deposit south of Hercules crater are dated as Upper Imbrian or older [Scott, 1972a; Grolier, 1974], and therefore >3.2 b.y. [Wilhelms, 1987]. No stratigraphic markers are available that can give a lower bound to the age of the cryptomafic units in this area. Therefore, the age of the Hercules cryptomafic deposit can be only roughly estimated as >3.2 b.y. old.


        Tsiolkovskiy
       Tsiolkovskiy crater (20º S, 128.5º E) is located on the western farside of the Moon. Several mare patches are located inside Tsiolkovskiy crater and on its ejecta deposits in the southwest [Wilhelms and El-Baz, 1977]. This ejecta obscures many old craters, some of which may contain cryptomafic materials.
       One dark-haloed crater is identified in Neujmin crater [Schultz and Spudis, 1979], which is located to the southwest of Tsiolkovskiy crater and covered by its ejecta [Wilhelms and El-Baz, 1977]. Topographic and crustal thickness lows (Figure 4) are found to the southwest of Tsiolkovskiy crater, where Neujmin crater is located. A slight FeO anomaly (Figure 5), relative to the surrounding area, is also associated with the Neujmin crater region and irregular patches of low albedo (Figure 6) can be seen in its vicinity. On the basis of these observations, a cryptomafic deposit, with an area of approximately 5.0 x 104 km2 is proposed for the region around Neujmin crater, to the southwest of Tsiolkovskiy. Adopting a conservative value of 400 meters for the thickness of cryptomafic materials, on the basis of average thicknesses of eastern maria [De Hon and Waskom, 1976], the volume of the Tsiolkovskiy cryptomafic deposit is estimated to be 2.0 x 104 km3.
       Tsiolkovskiy crater, whose ejecta overlies the cryptomafic materials, is dated as upper Imbrian [Wilhelms, 1987] and, therefore, approximately >3.2 b.y. old. Neujmin crater, which contains portions of the cryptomafic deposit, is dated as pre-Nectarian [Wilhelms, 1987] and, therefore roughly younger than 4.0 b.y. The age of the Tsiolkovskiy cryptomafic deposit can thus be estimated as roughly 3.2 - 4.0 b.y. old.


        Korolev
       The Korolev basin is an old, pre-Nectarian [Greeley et al., 1993b] basin on the central farside of the Moon. No mare units are found in Korolev, but numerous light plains deposits, of various ages, are located in and around the basin [Stuart-Alexander, 1978].
       Although no dark-haloed craters (Figure 2) are identified in the Korolev basin [Schultz and Spudis, 1979; 1983], geochemical anomalies have been note, particularly associated with the Imbrian aged light plains units [Hawke et al., 1985]. Clementine data indicates that a mascon is located in the basin region [Neumann et al., 1996], arguing for volcanic filling. Topography and crustal thickness maps (Figure 4) indicate that Korolev basin is located in a region where some of the highest elevations and crustal thicknesses on the Moon are found. However, a distinct region of low topography and crustal thinning can clearly be seen in this area. A minor albedo anomaly (Figure 6) is associated with the Korolev basin, but no FeO anomaly (Figure 5) is recognizable in this region. On the basis of the above observations a cryptomafic deposit associated with the light plains units in the Korolev area is proposed. The area of this possible cryptomafic unit is estimated to be 9.5 x 104 km2. For the thickness of the Korolev cryptomafic deposit, the average mare thickness (400 m) of the eastern and western maria [De Hon and Waskom, 1976; De Hon, 1979] is assumed. Multiplying this thickness by the areal estimate determined above give a volume of 3.8 x 104 km3 for the Korolev cryptomafic deposit.
       The Korolev basin has been dated at 4.04 b.y. [Greeley et al., 1993b]. The local light plains units have been dated at 4.02 b.y for the INp units, and 3.88 - <3.4 b.y. for the Ip units [Stuart-Alexander, 1978; Greeley et al., 1993b]. Using the age of the basin as the upper age bound and the youngest light plains age for the lower bound gives a rough estimate of 3.4 - 4.04 b.y. for the age of the Korolev cryptomafic deposit.


        Humboldtianum
       Humboldtianum is a complex, multi-ringed basin located in the northern areas of the eastern limb of the Moon. Several irregular mare deposits are found within the inner and outer rings of Humboldtianum basin and many lower Imbrian light plains units are similarly scattered throughout the region [Lucchitta, 1978].
       Galileo data has indicated the presence of mafic signatures associated with some of the light plains deposits [Belton et al., 1994]. Clementine data has demonstrated that the crust in Humboldtianum basin is anomalously thin (Figure 4), less than 40 km thick [Neumann et al., 1996]. A distinct FeO anomaly (Figure 5) is located in Humboldtianum, and albedo anomalies (Figure 6) are present inside the inner basin ring and in the southern regions between the inner and outer rings. Within the inner ring, the albedo anomalies appear to be related to the light plains units mapped by Lucchitta [1978]. In the southern section between rings, the albedo anomaly is not confined to the light plains units, but is also associated with Nectarian aged terra materials. The existence of two cryptomafic deposits is, therefore, proposed for Humboldtianum basin. A younger unit, associated with Imbrian aged light plains, may be located within the inner basin ring. An older unit, associated with Nectarian aged terra materials [Lucchitta, 1978], may be located in the southern region between the inner and outer rings. The areas of these two cryptomafic units are estimated to be approximately 1.12 x 105 km2 for the older unit and 2.1 x 104 km2 for the younger unit. A thickness of 400 meters is estimated for these cryptomafic deposits on the basis of average mare thicknesses [De Hon and Waskom, 1976; De Hon, 1979]. Multiplying this thickness by the areal estimates determined above yields a volume of 4.5 x 104 km3 for the cryptomafic deposit in the south of Humboldtianum basin and a volume of 8.4 x 103 km3 for the cryptomafic deposit in the center of Humboldtianum.
       The light plains units are dated as lower Imbrian in age [Lucchitta, 1978], and therefore >3.8 b.y. old [Wilhelms, 1987]. The Nectarian aged terra materials are approximately >3.87 b.y. old [Wilhelms, 1987]. The Humboldtianum basin is Nectarian in age and so roughly <3.92 b.y. old [Wilhelms, 1987]. For the younger cryptomafic unit, inside the inner ring of Humboldtianum, the terra material is adopted as the lower age bound and the light plains are adopted as the upper age bound. The age of this cryptomafic unit is, therefore, roughly 3.8 - 3.87 b.y. old. For the older cryptomafic unit, located in the southern regions between the inner and outer rings of Humboldtianum, the age can be roughly estimated as 3.87 - 3.92 b.y.


        Maurolycus
       Maurolycus crater (43º S, 15º E) is located in the south-central highlands on the nearside of the Moon. No mare materials are located in the vicinity of this crater, but a wide variety of light plains and terra materials mantle the region [Scott, 1972b].
       The presence of a small anomalous FeO area has been identified near Maurolycus crater [Giguere et al., 1998]. Also, several dark-rayed craters have been noted and related to the excavation of mafic materials [Giguere et al., 1998]. Several small topographic and crustal thickness anomalies (Figure 4) can be seen in the Maurolycus area, suggesting the presence of some small basins that have not been identified by Wilhelms [1987]. Also, a strong, roughly circular, albedo anomaly (Figure 6) is identified around Maurolycus crater and extending east. On the basis of these observations, the presence of a cryptomafic deposit surrounding Maurolycus crater is proposed. The area of this cryptomafic deposit is estimated to be 1.6 x 105 km2. Assuming a conservative thickness estimate of 400 meters, based on average mare thicknesses [De Hon and Waskom, 1976; De Hon, 1979], the volume of the Maurolycus cryptomafic deposit is estimated to be 6.4 x 104 km3.
       The units most closely associated with the albedo anomaly that defines the cryptomafic deposit have been dated as roughly upper Nectarian [Scott, 1972b], or >3.87 b.y. old [Wilhelms, 1987]. The basin that the Maurolycus cryptomafic deposit appears to occupy is not identified by Wilhelms [1987] and, therefor, no age is available for this basin. However, the albedo map (Figure 6) seems to indicate that Maurolycus crater ejecta is not superposed on top of the cryptomafic materials, so this craters is adopted as the lower stratigraphic marker for this cryptomafic deposit. Maurolycus crater has been dated as Nectarian, and thus <3.92 b.y. old [Wilhelms, 1987]. The age of the Maurolycus cryptomafic deposit is, therefore, estimated at roughly 3.87 - 3.92 b.y.


        Milne
       Milne is a small, pre-Nectarian basin [Wilhelms, 1987], located on the far-eastern side of the Moon between Australe basin and Tsiolkovskiy crater. No mare materials are located near Milne, but the basin floor is covered with light plains deposits [Wilhelms and El-Baz, 1977].
       Dark-haloed craters have been mapped in the center and north of the basin (Figure 2). Geochemical anomalies have been identified in the northwest, near Lacus Solitudinus [Hawke et al., 1985]. Minor topographic and crustal thickness lows are associated with the basin (Figure 4). A very minor FeO anomaly (Figure 5) is present in the Milne area. A prominent albedo low is evident in the albedo map of Figure 6. This albedo anomaly is distributed in a circle around the crater Scaliger, located on the northwestern limb of Milne, suggesting that Scaliger may itself be a dark-haloed crater.
       On the basis of these observations, the existence of a cryptomafic deposit in Milne and in regions north of the basin rim is suggested. The area of these deposits is estimated to be 9.2 x 104 km2. Because of the proximity of this cryptomafic area to the Australe basin, the thickness here is estimated on the basis of mare thickness calculations for Australe. Whitford-Stark [1979] estimated that mare materials in Australe range in thickness from 0.5 - 3.5 km. A conservative thickness estimate of 500 meters is, therefore, adopted for the Milne cryptomafic deposit. Multiplying this thickness by the areal estimate obtained above, the volume of cryptomafic material in and around Milne basin is estimated to be 4.6 x 104 km3.
       The Milne basin is pre-Nectarian in age [Wilhelms, 1987]. An age of <4.0 b.y is, therefore adopted for the lower age bound of the cryptomafic deposit in this area. The units that surround Milne basin are dated as being approximately the age of Nectaris basin, or slightly older [Wilhelms and El-Baz, 1977]. Therefore, the age of these units, roughly 3.92 b.y. [Wilhelms, 1987] is adopted for the upper age bound of the Milne cryptomafic deposits. The cryptomafic material in and around Milne crater can be roughly dated as 3.92 - 4.0 b.y. old.


        Mendeleev
       The Mendeleev basin is located on the far side of the Moon, more or less south of Mare Moscoviens. Mendeleev is a Nectarian aged basin, approximately 330 km in diameter [Wilhelms, 1987]. No mare deposits are found in the Mendeleev area, but several light plains units are located within the basin and outside the basin rim [Wilhelms and El-Baz, 1977; Stuart-Alexander, 1978].
       No dark-haloed craters are identified in the Mendeleev area (Figure 2), but high FeO and TiO2 values have been associated with the light plains units in eastern sections of the basin [Hawke et al., 1985]. Topographic and crustal thickness anomalies are present at the location of this basin (Figure 4), but no distinct albedo anomalies are associated with Mendeleev (Figure 6). On the basis of the geochemical evidence, a cryptomafic deposit is proposed for this region. The area of the Mendeleev cryptomafic deposit is estimated as 5.3 x 104 km2. The average mare thickness for maria on the Moon [De Hon and Waskom, 1976; De Hon, 1979] is considered, and so a thickness of 400 meters is adopted for the Mendeleev cryptomafic deposit. Multiplying this thickness by the areal estimate determined above yields a volume estimate of 2.1 x 104 km3 for the cryptomafic deposit in Mendeleev basin.
       The Mendeleev basin is Nectarian in age, and therefore <3.92 b.y old [Wilhelms, 1987]. The light plains units that define the Mendeleev cryptomafic deposit are dated as being Imbrian in age [Wilhelms and El-Baz, 1977; Stuart-Alexander, 1978] and therefore >3.2 b.y. old [Wilhelms, 1987]. The age of the Mendeleev cryptomafic deposit is thus roughly dated as 3.2 - 3.92 b.y. old.



Discussion

       Conservative estimates of the areas and volumes of cryptomafic deposits, determined in the previous section, are catalogue in Table 1. At the bottom of Table 1 are presented the totals of all cryptomafic regions that were considered . It should be remembered that the areas determined in the previous section were based on data sets that are expected to give minimum estimates. The areal estimates themselves, however, were very rough, with no details of the deposit boundary being taken into account. This may account for an overestimate of no more than 25% of cryptomafic area determinations. Volume determinations were always based on very conservative estimates of thicknesses. Thus, despite the possibility that areas of cryptomafic deposits may have been overestimated, it is likely that the volume estimates are minimum values. Also, the list of cryptomafic regions in Table 1 is not comprehensive. In all probability, many cryptomafic regions remain unidentified. Thus the total area and volume estimates of lunar cryptomafic areas, presented at the bottom of Table 1, are likely to be minimum estimates.
       The dark-haloed crater map of Schultz and Spudis [1979, 1983] (Figure 2) shows the presence of dark-haloed craters in regions that have not been recognized in the literature as containing cryptomafic deposits (Figure 1, Table 1). For example, the area north of central Mare Frigoris is located far west of the accepted Frigoris cryptomafic deposit [Greeley et al., 1993a; Belton et al., 1994; Mustard and Head, 1995] and the central highlands are located considerably north of the Maurolycus crater area. Yet both these areas contain dark-haloed craters, as well as a multitude of light plains deposits (Figure 3). These areas may harbor undiscovered cryptomafic deposits. Additional cryptomafic deposits may also be located in regions north of Mare Crisium, where several dark-haloed craters are identified (Figure 2), and south of Mare Moscoviens, where a single dark-haloed crater is mapped.
       The Clementine laser altimetry experiment has allowed the existence of several ancient basins to be identified or confirmed [Spudis et al., 1994]. Since volcanism occurs preferentially in basins [Head, 1975a], these new basins should be examined for potential cryptomafic materials. Such basin candidates include the ~500 km diameter Coulomb-Sarton basin, the ~700 km diameter Mutus-Vlacq basin, which is located southwest of Nectaris basin and which contains light plains deposits and possibly cryptomafic deposits, the 600 km Freudlich-Sharonov basin (18.5º N, 175º E), the 700 km Tsiolkovskiy-Stark basin (15º S, 128º E), and two basins northeast of Mare Moscoviens (30º N, 165º E, and 50º N, 165º E) measuring 330 and 450 km in diameter, respectively [Spudis et al., 1994]. A possible 300 km diameter basin at south pole has also been suggested [Shoemaker et al., 1994].
       Gravity modeling suggests that 3-6 km thicknesses of dense mafic materials may be present in the centers of circular maria [Bratt et al., 1985; Arkani-Hamed, 1998]. However, estimates of surface mare thicknesses yield maximum values in the range of only 1.5-2 km [De Hon and Waskom, 1976; De Hon, 1979; Staid et al., 1996]. This suggests that the surface mare units in circular maria, which range in age from 3.9-2.5 b.y. old [Head, 1976], may be underlain by Nectarian or pre-Nectarian ejecta deposits that obscure even older mafic units. This interpretation is consistent with the observation of several potentially multi-layered cryptomafic deposits during the course of this study, as well as the suggestion that interleaving of ejecta deposits between volcanic layers may be a common phenomenon. An age of >3.9 b.y. for such sub-mare deposits is also consistent with the ancient age range of many of the cryptomafic deposits identified in this study (Table 1). Thus, unidentified ancient mafic deposits may underlie many of the known basin maria.
       Clearly, the existence of any of the above-mentioned cryptomafic deposits would affect the area and volume totals that are calculated in Table 1. However, to assess the impact of such units, detailed evaluations are needed just to confirm the existence of these proposed cryptomafic deposits. Such a study is well beyond the scope of this thesis. The survey of proposed cryptomafic deposits that was conducted in the previous section (Figure 1, Table 1) probably yields minimum values for the global area and volume estimates of cryptomafic deposits, providing a good beginning for the study of their importance and their affect on our understanding of global lunar volcanism.


        Surface Areas of Extrusive Volcanism
       The total surface area of all cryptomafic deposits identified by this study may represent as much as be 3.6 x 106 km2 (Table 1). The validity of these areal estimates can be assessed by considering the results of Greeley et al. [1993a]. On the basis of Galileo Earth-Moon 2 flyby data, cryptomafic deposits northeast of Frigoris, east of Marginis, and southwest of Smythii (the Balmer basin region in this study) are estimated to represent a total area of ~ 6 x 105 km2 [Greeley et al., 1993a]. From Table 1, the total for these three areas from this study can be seen to be ~ 6.4 x 105 km2. The areal estimates from this study are, therefore, found to be comparable to earlier rough estimates, validating the results and techniques of this study.
       Known maria represent a total surface area of approximately 6.3 x 106 km2 [Head, 1975b]. Cryptomafic deposits may have a total surface area of 3.6 x 106 km2 (Table 1). However, in the Australe and Smythii cryptomafic regions, areal determinations included suspected sub-mare deposits. To determine the total surface area of all mafic units, the contributions from these sub-mafic deposits must be removed so as not to count them twice. Mare deposits, in the regions in question, correspond to a total area of 0.4 x 106 km2. Thus, the total surface area of the Moon that may have been covered by extrusive deposits at some point in its history is now estimated to be 9.5 x 106 km2 and cryptomafic deposits represent a potential increase of 50% to the surface area of known mafic materials. In comparison, the surface area of the Moon is equal to 37.9 x 106 km2 [Heiken et al., 1991]. Therefore, based on the estimates of cryptomafic deposits from Table 1 and previously published mare estimates [Head, 1975b], 25% of the surface area of the Moon may have been covered by the products of extrusive volcanism at one time or another.


        Volumes of Extrusive Volcanism
       The total volume of all cryptomafic deposits identified by this study is estimated to be approximately 1.6 x 106 km3 (Table 1). In comparison, a volume of 10 x 106 km3 has been estimated for all of the known lunar maria [Head, 1975b]. However, concern has been expressed that this estimate may be more than a factor of 2 too high [Hörz, 1978; Budney and Lucey, 1998]. Hörz [1978] has even suggested that the total volume of mare materials may be as low as 1-2 x 106 km3. Therefore, a re-evaluation of known mare volumes is required before the contributions of mare and cryptomafic materials can be compared.
       The mare volume estimates of Head [1975b] are based on the thickness estimates of De Hon [1974], and therefore suffer from several limitations. Firstly, these thickness estimates are preliminary estimates that were later updated by De Hon and Waskom [1976] and De Hon [1979]. The volume estimates of Head [1975b], therefore, assume thicknesses on the order of 1000 meters for shallow maria and maximum thicknesses of 8-10 km in the centers of circular maria, instead of values on the order of ~400 meters and ~1500 meters, respectively [De Hon and Waskom, 1976; De Hon, 1979]. Secondly, the techniques used by De Hon [1974], as well as De Hon and Waskom [1976] and De Hon [1979], for estimating mare thickness have been noted to give erroneously high values [Hörz,1978; Budney and Lucey, 1998]. In early estimates, the thickness of mare deposits was determined primarily by considering the rim heights of partially filled craters and comparing them to theoretically determined rim heights [De Hon, 1974, 1979; De Hon and Waskom, 1976]. However, this technique assumes that all flooded craters are fresh. Hörz [1978] suggests that a model based on degraded craters may be more appropriate and since the rim height of degraded craters is ½ that of fresh craters, mare thicknesses estimated by this technique may be more than a factor of 2 too high. Also, this technique does not consider the possibility of filling by ejecta from adjacent craters [De Hon, 1974; Hörz, 1978]. As was observed in Chapter 3 and in several cryptomafic deposits in the previous section, interleaving of ejecta deposits between volcanic layers may be a common phenomenon. Thus, techniques which measure crater fill may be including ejecta deposits in their mare thickness calculations and so producing overestimated values.
       Techniques for estimating mare thicknesses, which do not suffer from the same limitations as the De Hon [1974] method, are available. A variety of methods for estimating mare thicknesses are summarized by Yingst and Head [1997]. Of the methods they describe, the measurement of post-mare crater penetration and flow lobe height are two techniques that do not depend on a knowledge of pre-fill structures. Crater penetration was used by Budney and Lucey [1998] to estimate the thickness of materials in Mare Humorum. The thickness of this mare was found to range from 200 - 500 meters [Budney and Lucey, 1998], agreeing with the estimates of Hörz [1978] that were based on filling of degraded craters. Crater penetration was also used by Staid et al. [1996] to estimate the thickness of Mare Tranquillitatis. Here, thicknesses were found to be ~2 km in the mare center and ~600 meters towards the mare edges [Staid et al., 1996]. These results suggest that the thicknesses used by Head [1975b] to estimate mare volumes were much too high.
       Mare thickness estimates have also been obtained from gravity modeling of lunar mascons [Solomon and Head, 1980]. Thicknesses values at the centers of large circular maria were found to be on the order of 4.5 km in Maria Smythii and Nectaris, 3.6 km in Grimaldi, 1.7 km in Orientale, >7km in Crisium, >9 km in Imbrium, 2.7 km in Humorum, and 8.5 km in Serenitatis. Thicknesses at mare edges were found to be at least 1 km for all basins except Humorum, where mare thicknesses of only 500 meters were found at the mare edges [Solomon and Head, 1980]. However, later gravity modeling yields thicknesses of 0.5-1.3 km for the irregular maria, and 3.5-4.5 km in the centers of circular maria [Bratt et al., 1985]. Furthermore, mascon modeling of Clementine gravity data yields thickness that are a maximum of 3-6 km at the centers of circular mare [Arkani-Hamed, 1998]. These later results are consistent with the interpretation that the thicknesses used by Head [1975b] to estimate global mare volumes were much too high.
       On the basis of the above considerations, it is estimated that the mare volumes of Head [1975b] are overestimated by a factor of at least 2. Thus, the volume of known maria can be estimated as no more than 5 x 106 km3. Recalling that the total volume of cryptomafic deposits was estimated to be 1.6 x 106 km3 (Table 1), the cryptomafic deposits of this study may, therefore, represent a 32% increase in the total volume of mafic materials on the Moon. The total volume of all mafic deposits on the Moon may be potentially as much as 6.6 x 106 km3.


        Volcanic Fluxes
       Head and Wilson [1992] assessed the lunar volcanic flux, as a function of time, by considering the total volume of lava extruded during a particular time period and assuming it was evenly distributed over that time. In this manner, they estimate the volcanic flux per annum to be 0.015 km3/a during the Upper Imbrian epoch (3.8-3.2 b.y.), 1.3 x 10-4 km3/a in the Eratosthenian period (3.2-1.1 b.y.), and 2.4 x 10-6 km3/a in the Copernican period (1.1 b.y. - present). However, an average mare thickness of 1.5 km was used for the calculations of flux in the Upper Imbrian epoch [Head and Wilson, 1992]. As was noted above, this value is at least a factor of 2 too large for an average mare thickness estimate. Thus, a more appropriate value of volcanic flux during the Upper Imbrian period would be ~7 x 10-3 km3/a, or half of the value estimated by Head and Wilson [1992].
        From Table 1, the volume of all cryptomafic units that can be decidedly and unambiguously dated as older than 3.8 b.y. can be calculated. The results indicate that a potential volume of 13.5 x 105 km3 can be roughly dated in the age range of 3.8 - 4.1 b.y. Following the example of Head and Wilson [1992], it is assumed that the volcanism represented by these cryptomafic deposits was evenly distributed over the age range determined above. In this manner, the flux of these ancient mafic materials is found to be ~4.5 x 10-3 km3/a. Comparing this flux to that obtained for the Upper Imbrian period above (~7 x 10-3 km3/a), it is clear that the lunar volcanic flux prior to 3.8 b.y. ago may have been a significant fraction (60 %) of to the flux after 3.8 b.y. Thus, volcanism prior to 3.8 b.y may have been highly significant, and possibly even comparable to volcanism in the late Imbrian period, when the volcanic flux is believed to have peaked [Head and Wilson, 1992].
       If volcanism was so significant prior to 3.8 b.y., it could be argued that young impact basins, such as Imbrium and Orientale, should have excavated plutonic material from the dikes that fed these cryptomafic deposits. However, no plutonic mare materials were identified in the returned lunar samples [i.e., Heiken et al., 1991]. This could be explained by several possibilities. First, the volcanic flux could be concentrated at 3.8 b.y. The ages identified in this study give only ranges of possible ages. Very few deposits have upper age bounds greater than 3.8 b.y. Thus cryptomafic deposits may have been emplaced at the youngest extreme of the range given, with very few being emplaced much earlier than 3.8 b.y. Second, cryptomafic deposits may not have been located at the locations of the youngest impacts, thus no plutonic materials would have been excavated. Third, the early volcanic materials may not be mare basalts, but instead may have consisted of other mafic materials. In this case, the plutonic materials in question could correspond to the Mg-rich suit of lunar rocks. Such a finding would have important implications for the petrogenesis of lunar volcanic materials. Fourth, the volume represented by intrusive dikes may be very insignificant when compared to the crustal volume. In this case, even if plutonic dike material had been excavated, it would be sufficiently rare that its absence from the sample record would be statistically unremarkable. Fifth, examples of plutonic materials may only be present in the form of melt clasts. Plutonic feeder dikes are expected to occur at greater depths than the surface deposits. Impact melt materials tends to derive from greater depths in the crater than solid ejecta materials [i.e., Melosh, 1989], thus deeper materials are expected to be excavated as impact melt, loosing their pre-impact texture. Therefore, the absence of plutonic mare-basalt clasts in the lunar record does not necessarily argue against extensive early volcanism.


        Implications for Lunar Evolution Models
       This study of global lunar cryptomafic deposits has shown that these deposits may represent a significant component of lunar volcanism, increasing our understanding of the volume and flux of early volcanic deposits. These findings are bound to affect current lunar evolutions models [i.e., Solomon and Head, 1979; Neal and Taylor, 1992; Hess and Parmentier, 1995] and possible implications are, therefore, briefly discussed here.
       The results of this study suggest that mafic materials have been extruded on the lunar surface prior to 3.9 b.y. ago and possibly as long ago as 4.1 b.y. (Table 1). In comparison, surface mare ages range from 3.9 - 2.5 b.y [Head, 1976; Solomon and Head, 1980]. Thus, the onset of volcanism may have occurred considerably earlier than is indicated by the surface maria. Furthermore, the high volume and flux of ancient cryptomafic deposits infers that volcanism may have been very significant in the early periods of lunar history. Previously, the existence of ancient basalts was known, but only from rare basaltic clasts in returned lunar samples [i.e., Taylor et al., 1983; Nyquist and Shih, 1992] and the volumetric significance of such ancient basalts was not known [Head and Wilson, 1992]. The results of this study suggest that volcanism may have been very prominent in the period between 4.1 - 3.8 b.y., possibly comparable to volcanism in the late Imbrian period.
       The possibility of extensive volcanism 4.1 - 3.8 b.y. ago can provide important constraints on lunar evolution models. While the results of this study cannot distinguish between the various models, they can be used to help constrain model parameters. Evolution models can be categorized into two basic types, based on thermal or geochemical considerations. For thermal models, the presence of extensive volcanism prior to 3.8 b.y. suggests that the lunar interior may have been hotter than previously recognized during its early history. For geochemical models, the possibility of extensive volcanism as early as 4.1 b.y. places constraints on the timing of events. Namely, the melt source regions must be emplaced well before 4.1 b.y. to account for the possibility of early and plentiful lavas.

        Magma Ocean Models
       Both thermal and geochemical constraints require the existence of some kind of magma ocean in the early history of the Moon [i.e., Solomon and Chaiken, 1976; Nyquist and Shih, 1992]. Depths on the order of 200-300 km [Solomon and Chaiken, 1976; Solomon, 1980] to 630 km [Kirk and Stevenson, 1989] have been proposed for the lunar magma ocean. Even depths as great as 1000 km have been justified by invoking a slushy "magmifer", where the molten outer layer is not completely liquid [Warren, 1985].
       In any case, geochemical considerations indicate that basalt lavas are formed by partial melting [i.e., Taylor et al., 1971] and so cannot be the liquid byproducts of a magma ocean. Thus, the magma ocean must cool and crystallize in order to provide the source regions of basaltic magmas. This study suggests that the magma ocean must cool by 4.1 b.y. in order to account for the early mafic lavas that formed the very ancient cryptomafic deposits. Most models estimate that the magma ocean cools very quickly. Solomon and Longhi [1977] suggest that a 400 km deep magma ocean should be completely crystallized by 4.4 b.y. Nyquist et al. [1977] propose that crystallization of the magma ocean proceeds downwards from the base of the crust, thus crystallized source regions should be available very early, at least at the base of the crust. A similar explanation is invoked for the "magmifer" model, where the slushy layer migrates downwards over time, leaving behind crystallized source regions [Shirley, 1983].
       Thus, it would appear that the crystallization of the lunar magma ocean is not constrained by the results of this study. However, the potential for extensive early volcanism argues for a hotter early interior for the Moon. More initial heat may mean a deeper initial magma ocean [Hood, 1986].

        Thermal Models
       A popular thermal evolution model for the Moon is the model of Solomon and Head [1979, 1980], where partial melting through time is modeled on the basis of global and local stress constraints (Figure 7). In this model, cessation of volcanism was caused by the switch to compressive stresses in the crust, not from lack of melt production, and the volume of the Moon has remained relatively constant over the past 3.8 b.y. [Solomon and Chaiken, 1976]. Thermal expansion of the Moon is confined to the 1st billion years of lunar history, thus the heat peak must occur prior to 3.6 b.y. ago [Solomon and Head, 1980] and the peak of volcanic flux must also occur in this time frame [Head and Wilson, 1992].
       The results of this study are consistent with this thermal model. The peak flux of extrusive volcanism occurs in the Imbrian period [Head and Wilson, 1992], thus coinciding with the heat peak and the time of extensive stresses [Solomon and Head, 1979, 1980]. The presence of extensive early volcanism would argue for a hotter early interior for the Moon than was previously supposed. However, Solomon and Head [1980] state that warmer initial temperature profiles are still consistent with the tectonic relationships of the surface maria, but much hotter models are excluded on the basis of the large compressive stresses that would result at later times. Thus, extensive early volcanism is still consistent with the thermal model of Solomon and Head [1979, 1980].

        Geochemical Models
       Petrologic considerations require basalts to be derived by partial melting [Taylor et al., 1971] of source regions at depths of 300-400 km [Ringwood and Essene, 1970; Walker et al., 1974]. Despite the potential contributions of cryptomafic deposits, as estimates by this study, the total volume of extrusive volcanism is still relatively low, thus only very limited partial melting of the source regions is required [Hörz, 1978].
        Neal and Taylor [1992] classified the various models of basalt formation into four basic types; primitive source, assimilation, cumulate layer cake, and cumulate overturn. Primitive source models call for basalt formation from the melting of primitive unfractionated lunar material [i.e., Ringwood and Essene, 1970]. Assimilation models incorporate late stage cumulates from the magma ocean into the melt [i.e., Kesson and Ringwood, 1976]. Cumulate layer cake models require the melting of shallow depth [Ma et al., 1976] olivine and pyroxene cumulates [Nyquist et al., 1977] from the magma ocean, which then mix with melts of primitive materials [Taylor and Jakeû, 1974; Solomon and Longhi, 1977; Papike and Vaniman, 1978]. Deepening of the melting regime within the stratified source region is supposed to account for the age and composition variations in observed lunar basalts [i.e., Taylor and Jakeû, 1974]. Cumulate overturn models propose that a gravitational overturn of the magma ocean-produced cumulate pile results in mixing of cumulates and primitive mantle sources [i.e., Hughes et al., 1989]. Neal and Taylor [1992] favor the overturn models, which have been shown to be most compatible with the isotopic data [Nyquist and Shih, 1992].
       A recent model of cumulate overturn was considered by Hess and Parmentier [1994; 1995]. In their model, the cumulates are shown to be gravitationally unstable with respect to the underlying primitive mantle, causing a density-driven overturn [Hess and Parmentier, 1994]. The dense ilmenite cumulates collect at the base of this overturn, heating the overlying primitive mantle material, forming melts that coalesce into unstable plumes and rise convectively [Hess and Parmentier, 1994]. A variety of model parameters (i.e., heating rate, viscosity, and density gradient) yield differing times for initial melt formation, where lower density gradients, lower radioactivity, and increased viscosity result in later initial melting [Hess and Parmentier, 1995; Parmentier and Hess, 1995]. Initial melt times, ranging from 500-700 m.y. after overturn, with extreme values of ~1000, 2000 m.y., were found [Hess and Parmentier, 1995].
       The results of this study indicate that relatively large quantities of melt formation may be required as early as 4.1 b.y. ago, or 500 m.y. after the formation of the Moon. This suggests that, unless magma ocean formation and cumulate overturned occur very quickly (<100 m.y.), the model of Hess and Parmentier [1994, 1995] may not be able to account for the formation of the earliest lavas. Spera [1992] suggest that the buildup of cumulate instability and overturn takes ~400 m.y., thus much shorter melt generation times would be required to produce the early mafic deposits observed. This places very important constraints on the parameters of the Hess and Parmentier [1994, 1995] model, namely the heating rate, viscosity, and density gradient of the early lunar interior. However, it has been argued that overturn causes decompression melting [Spera, 1992]. It is not clear if this decompression melting would be sufficient or timely enough to account for the earliest cryptomafic deposits in this study. Clearly, studies of this nature can be used to place important constraints on overturn models.
       It should also be noted that the results of this study suggest that extrusive volcanism was extremely insignificant on the Moon. Despite a potential increase of 30% in volcanic products, the total volume of extrusive mafic material still appears to be very low, representing an insignificant fraction of the volume of the potential source regions. Thus the total volume of melt produced is expected to be equally low, arguing for a source region that is difficult to melt. This suggests that the cumulate products of the Magma Ocean may be very difficult to re-melt [Hess and Parmentier, 1995]. Conversely, intrusion/extrusion ratios have been suggested to be potentially as high as 50:1 [Head and Wilson, 1992]. In this case, large increases in extrusive volumes would have a profound effect on the total volume of melt produced, suggesting that melting of lunar source regions may not be as difficult as suggested.



Conclusions

       Rough estimates of volumes and areas have been determined for all known or suspected cryptomafic deposits on the Moon, proposing a minimum global estimate of lunar cryptomafic materials. These areal and volume estimates should not be considered as comprehensive, since many other cryptomafic deposits may still be undiscovered, in unstudied areas, in unrecognized basins, or hidden beneath the surfaces of known mare deposits. Much work is required to recognize and catalogue such cryptomafic units and future work should strive to apply the techniques that were developed in previous chapters of this thesis to study these, as well as previously documented, cryptomafic deposits in greater detail.
       In the mean time, the results of this study have shown that cryptomafic deposits on the Moon may be extremely significant. Identified cryptomafic regions have a potential total area of 3.6 x 106 km2, representing ~50% of the surface area of mare deposits, and a conservative total volume of 1.6 x 106 km3. Mare volumes were recalculated to be ~5 x 106 km3, thus cryptomafic deposits may correspond to ~30% of known mare volumes. On the basis of these results, extrusive volcanism is suggested to represent a total area of 9.5 x 106 km2, corresponding to 25% of the surface area of the Moon, and a total volume of 6.6 x 106 km3. The fluxes of early (>3.8 b.y. old) volcanics were found to be ~4.5 x 10-3 km3/a, approximately 60% of the fluxes in the late Imbrian period (3.8-3.2 b.y.).
       The implications of these results is that volcanism may have begun much sooner than had previously been recognized and may have been much more voluminous during the early period of lunar history. Such findings suggest that the early lunar interior may have been hotter than previously believed and that extensive melting of source regions may have begun at an earlier time. These results place important constraints on models of lunar evolution.



References

Arkani-Hamed, J., The lunar mascons revisited, J. Geophys. Res., 103, 3709-3739, 1998.

Andre C.G., R.W. Wolfe, I. Adler, and P.E. Clark, Are early magnesium-rich basalts widespread on the Moon?, Proc. Lunar Planet. Sci. Conf. 10th. 1739 - 1751, 1979.

Antonenko, I., J.W. Head, J.F. Mustard, and B.R. Hawke, Criteria for the detection of lunar cryptomaria, Earth, Moon, Planets, 69, 141-172, 1995.

Bell, J.F. and B.R. Hawke, Lunar Dark-haloed impact craters: Origins and implications for early mare volcanism, J. Geophys. Res., 89, 6899-6910, 1984.

Belton, M.S., J.W. Head, C.M. Pieters, R. Greeley, A.S. McEwen, G. Neukum, K.P. Klaasen, C.D. Anger, M.H. Carr, C.R. Chapman, M.E. Davies, F.P. Fanale, P.J. Gierasch, R. Greenberg, A.P. Ingersoll, T. Johnson, B. Paczkowski, C.B. Pilcher, and J. Veverka, Lunar impact basins and crustal heterogeneity: New western limb and far side data from Galileo, Science, 255, 570-576, 1992.

Belton, M.S., R. Greeley, R. Greenberg, A.S. McEwen, K.P. Klaasen, J.W. Head, C.M. Pieters, G. Neukum, C.R. Chapman, P.Geissler, C. Heffernan, H. Breneman, C.D. Anger, M.H. Carr, M.E. Davies, F.P. Fanale, P.J. Gierasch, A.P. Ingersoll, T.V. Johnson, C.B. Pilcher, W.R. Thompson, J. Veverka, and C. Sagan, Galileo multispectral imaging of the north polar and eastern limb regions of the Moon, Science, 264, 1112-1115, 1994.

Blewett, D.T., B.R. Hawke, and P.G. Lucey, Exploring the Crisium region with multispectral imagery, Lunar Planet. Sci. Conf., XXVI, 137-138, 1995a.

Blewett, D.T., B.R. Hawke, P.G. Lucey, G.J. Taylor, R. Jaumann, and P.D. Spudis, Remote sensing and geologic studies of the Schiller-Schickard region of the Moon, J. Geophys. Res., 100, 16,959-16,978, 1995b.

Blewett, D.T., P.G. Lucey, and B.Ray Hawke, Clementine images of lunar sample-return stations: Refinement of FeO and TiO2 mapping techniques, J. Geophys. Res., 102, 16,319-16,325, 1997.

Bratt, S.R., S.C. Solomon, J.W. Head, and C.H. Thurber, The deep structure of lunar basins: Implications for basin formation and modification, J. Geophys. Res., 90, 3049-3064, 1985.

Budney, C.J. and P.G. Lucey, Basalt thickness in Mare Humorum: The crater excavation method, Submitted to J. Geophys. Res., 1998.

Chao, E.C.T., L.A. Soderblom, J.M. Boyce, D.E. Wilhelms, and C.A. Hodges, Lunar Light Plains Deposits (Cayley Formation) - A Reinterpretation of Origin, Lunar Planet. Sci. Conf., IV, 127-128, 1973.

Clark, P.E. and B.R. Hawke, The relationship between geology and geochemistry in the Undarum/Spumans/Balmer region of the Moon, Earth, Moon, and Planets, 38, 97-112, 1987.

Clark, P.E. and B.R. Hawke, The lunar farside: The nature of highlands east of Mare Smythii, Earth, Moon, and Planets, 53, 93-107, 1991.

Coombs, C.R. and B.R. Hawke, Pyroclastic deposits on the western limb of the Moon, Proc. Lunar Planet. Sci. Conf. 22nd, 303-312, 1992.

Conca J. and N. Hubbard, Evidence for early volcanism in Mare Smythii, Proc. Lunar Planet. Sci. Conf. 10th, 1727-1737, 1979.

De Hon, R.A., Thickness of mare material in the Tranquillitatis and Nectaris basins, Proc. Lunar Planet. Sci. Conf. 5th, 53-59, 1974.

De Hon, R.A., Thickness of the western mare basalts, Proc. Lunar Planet. Sci. Conf., 10th, 2935-2955, 1979.

De Hon, R.A. and J.D. Waskom, Geologic structure of the eastern mare basins, Proc. Lunar Planet. Sci. Conf., 7th, 2729-2746, 1976.

Frontispiece, Proc. Lunar Sci. Conf. 8th, Lunar and Planetary Institute, Houston, TX, 1977.

Frontispiece, Proc. Lunar Sci. Conf. 9th, Lunar and Planetary Institute, Houston, TX, 1978.

Giguere, T.A., B.R. Hawke, G.J. Taylor, and P.G. Lucey, Geochemical studies of lunar cryptomare, Lunar and Planet. Sci. Conf., XXIX, Abstract #1782, Lunar and Planetary Institute, Houston, (CD-ROM), 1998.

Greeley, R., M.J.S. Belton, J.W. Head, A.S. McEwen, C.M. Pieters, G. Neukum, T.L. Becker, E.M. Fischer, S.D. Kadel, M.S. Robinson, R.J. Sullivan, J.M. Sunshine, and D.A. Williams, Galileo imaging results from the second Earth-Moon flyby: Lunar maria and related units, Lunar Planet. Sci. Conf., XXIV, 565-566, 1993a.

Greeley, R., S.D. Kadel, D.A. Williams, L.R. Gaddis, J.W. Head, A.S. McEwen, S.L. Murchie, E. Nagel, G. Neukum, C.M. Pieters, J.M. Sunshine, R. Wagner, and M.J.S. Belton, Galileo Imaging Observations of Lunar Maria and Related Deposits, J. Geophys. Res. 98, 17,183-17,205, 1993b.

Grolier, M.J., Geologic Map of the Geminus Quadrangle of the Moon, scale 1:1,000,000, Map I-841, USGS, Department of the Interior, Washington, DC, 1974.

Hawke, B.R., Geochemical anomalies on the lunar eastern limb and farside, Conf. Lunar Highlands Crust Lunar and Planetary Institute, Houston, TX, 56-58, 1979.

Hawke, B.R. and J.F. Bell, Remote sensing studies of lunar dark-halo impact craters: Preliminary results and implications for early volcanism, Proc. Lunar Planet. Sci. Conf., 12th, 665-678, 1981.

Hawke, B.R. and D.T. Blewett, Spectral and radar studies of the Schiller-Schickard region of the Moon, Lunar Plant. Sci. Conf., XXV, 515-516, 1994.

Hawke, B.R. and P.D. Spudis, Geochemical anomalies on the eastern limb and farside of the Moon, in Proc. Conf. Lunar Highlands Crust, J.J. Papike and R.B. Merrill (eds.), Pergamon, New York, N.Y., 467-481, 1980.

Hawke, B.R., P.D. Spudis, and P.E. Clark, The origin of selected lunar geochemical anomalies: Implications for early volcanism and the formation of light plains, Earth, Moon, and Planets 32, 257-273, 1985.

Hawke, B.R., C.A. Peterson, P.G. Lucey, G.J. Taylor, D.T. Blewett, B.A. Campbell, C.R. Coombs, and P.D. Spudis, Remote sensing studies of the terrain northwest of Humorum basin, Geophys. Res. Lett., 20, 419-422, 1993.

Head, J.W., Small-scale analogs of the Cayley formation and Descartes mountains in impact associated deposits, in Apollo 16 Preliminary Science Report, NASA, Houston, 27.16-27.20, 1972.

Head, J.W., Mode of occurrence and style of emplacement of lunar mare deposits, in Origins of Mare Basalts and their Implications for Lunar Evolution, Lunar Science Institute, Houston, TX, 61-65, 1975a.

Head, J.W., Lunar mare deposits: Areas, volumes, sequence, and implication for melting in source areas, in Origins of Mare Basalts and their Implications for Lunar Evolution, Lunar Science Institute, Houston, TX, 66-69, 1975b.

Head, J.W., Lunar volcanism in space and time, Rev. Geophys. and Space Phys., 14, 265-300, 1976.

Head, J.W. and L. Wilson, Lunar mare volcanism: Stratigraphy, eruption conditions, and the evolution of secondary crusts, Geochim. et Cosmochim. Acta, 56, 2155-2175, 1992.

Head, J.W., M. Belton, R. Greeley, C. Pieters, E. Fischer, J. Sunshine, K. Klaasen, A. McEwen, T. Becker, G. Neukum, J. Oberst, C. Pilcher, J. Plutchak, M. Robinson, T. Johnson, D. Williams, S. Kadel, R. Sullivan, I. Antonenko, N. Bridges, Lunar impact basins: New data for the nearside northern high latitudes and eastern limb from the second Galileo flyby, Lunar Planet. Sci. Conf. XXIV, 623-624, 1993a.

Head, J.W., S.M. Murchie, J.F. Mustard, C.M. Pieters, G. Neukum, A.S. McEwen, R.F. Greeley, E. Nagel, and M.J.S. Belton, Lunar impact basins: New data for the western limb and far side (Orientale and South Pole-Aitken basins) from the first Galileo flyby, J. Geophys. Res., 98, 17,149-17,181, 1993b.

Heiken, G.H., D.T. Vaniman, B.M. French, Lunar Sourcebook: A User's Guide to the Moon, Cambridge University Press, New York, NY, p736, 1991.

Hess, P.C. and E.M. Parmentier, Overturn of magma ocean ilmenite cumulate layer (II): Implications for lunar thermal and magmatic evolution, Lunar Planet. Sci., XXV, 541-542, 1994.

Hess, P.C. and E.M. Parmentier, A model for the thermal and chemical evolution of the Moon's interior: Implications for the onset of mare volcanism, Earth and Planet. Sci. Lett. , 134, 501-514, 1995.

Hood, L.L., Geophysical constraints on the lunar interior, in Origin of the Moon, LPI, Houston, W.K. Hartmann, R.J. Phillips, G.J. Taylor (eds.), 361-410, 1986.

Hörz, F., How thick are lunar mare basalts? Proc. Lunar Planet. Sci. Conf. 9th, 3311-3331, 1978.

Howard, K.A., D.E. Wilhelms, and D.H. Scott, Lunar Basin Formation and Highland Stratigraphy, Rev. Geophys. and Space Phys. 12 (3), 309-327, 1974.

Hughes, S.S., J.W. Delano, and R.A. Schmitt, Petrogenetic modelling of 74220 high-Ti orange volcanic glasses and the Apollo 11 and 17 high-Ti mare basalts, Proc. Lunar Planet Sci. Conf. 19-th, 175-188, 1989.

Kesson, S.E. and A.E. Ringwood, Mare basalt petrogenesis in a dynamic Moon, Earth Planet. Sci. Lett. , 30, 133-156, 1976.

Kirk, R.L. and D.L. Stevenson, The competition between thermal contraction and differentiation in the stress history of the Moon, J. Geophys. Res., 94, 12,133-12,144, 1989.

Lucchitta, B.K., Geologic Map of the North Side of the Moon, scale 1:5,000,000, Map I-1062, USGS, Department of the Interior, Washington, DC, 1978.

Lucey, P.G., G.J. Taylor, and E. Malaret, Abundance and distribution of iron on the Moon, Science, 268, 1150-1153, 1995.

Lucey, P.G., D.T. Blewett, and B.Ray Hawke, Mapping the FeO and TiO2 content of the lunar surface with multispectral imagery, J. Geophys. Res., 103, 3679-3699, 1998a.

Lucey, P.G., G.J. Taylor, B.Ray Hawke, and P.D. Spudis, FeO and TiO2 concentrations in the South Pole-Aitken basin: Implications for mantle composition and formation, J. Geophys. Res., 103, 3701-3708, 1998b.

Ma, M.-S., A.V. Murali, and R.A. Schmitt, Chemical constraints for mare basalt genesis, Proc. Lunar Sci. Conf. 7th, 1673-1695, 1976.

Maxwell, T.A. and C.G. Andre, The Balmer Basin: Regional Geology and Geochemistry of an Ancient Lunar Impact Basin, Proc. Lunar Planet. Sci 12th, 715-725, 1981.

Maxwell, T.A. and F. El-Baz, The nature of rays and sources of highland material in Mare Crisium, in Mare Crisium: The view from Luna 24, R.B. Merrill and J.J. Papike (eds.), Pergamon, N.Y., 89-103, 1978.

McCauley, J.F., Geologic map of the Hevelius region of the Moon, scale 1:1,000,000, Map I-491, USGS, Department of the Interior, Washington, DC, 1967.

McEwen, A.S., L.R. Gaddis, G. Neudum, H. Hoffman, C.M. Pieters, and J.W. Head, Galileo Observations of post-Imbrium lunar craters during the first Earth-Moon flyby, J. Geophys. Res., 98, 17,207-17,231, 1993.

McGetchin, T.R., M. Settle, and J.W. Head, Radial thickness variation in impact crater ejecta: implications for lunar basin deposits, Earth Planet. Sci. Letters, 20, 226-236, 1973.

Melosh, H.J., Impact Cratering: A Geologic Process, 245pp, Oxford University Press, New York, NY, 1989.

Metzger, A.E., J.I. Trombka, L.E. Peterson, J.C. Reed, and J.R. Arnold, Lunar surface radioactivity: Preliminary results of the Apollo 15 and Apollo 16 gamma-ray spectrometer experiments, Science, 179, 800-803, 1973.

Metzger, A.E., J.I. Trombka, R.C. Reedy, and J.R. Arnold, Element concentrations from lunar orbital gamma-ray measurements, Geochim. Cosmochim. Acta, suppl. 5, 2, 1067-1078, frontispiece, pl. 2, 1974.

Mustard, J.F., and J.W. Head, Evidence for cryptomare in the northern light plains of the Moon, Lunar Planet. Sci. Conf., XXVI, 1023-1024, 1995.

Mustard, J.F., and J.W. Head, Buried stratigraphic relationships along the southwestern shores of Oceanus Procellarum: Implications for early lunar volcanism, J. Geophys. Res., 101, 18,913-18,925, 1996.

Mustard, J.F., J.W. Head, S.M. Murchie, C.M. Pieters, M.S. Belton, and A.S. McEwen, Schickard Cryptomare: Interaction between Orientale ejecta and pre-basin mare from spectral mixture analysis of Galileo SSI data, Lunar and Planet. Sci. Conf., XXIII, 957-958, 1992.

Mustard, J.F., J.W. Head, and I. Antonenko, Mare-highland mixing relationships along the southwestern shores of Oceanus Procellarum, Lunar and Planet. Sci. Conf. , 24, 963-964, 1994.

Neal, C.R. and L.A. Taylor, Petrogenesis of mare basalts: A record of lunar volcanism, Geochim. et Cosmochim. Acta, 56, 2177-2212, 1992.

Neumann, G.A., M.T. Zuber, D.E. Smith, and F.G. Lemoine, The lunar crust: Global structure and signature of major basins, J. Geophys. Res., 101, 16,842-16,863, 1996.

Nyquist, L.W. and C.-Y. Shih, The Isotopic Record of Lunar Volcanism, Geochim. et Cosmochim. Acta 56, 2213-2234, 1992.

Nyquist, L.E., B.M. Bansal, J.L. Wooden, and H. Wiesmann, Sr-isotopic constraints on the petrogenesis of Apollo 12 mare basalts, Proc. Lunar Sci. Conf. 8th, 1383-1415, 1977.

Papike, J.J. and D.T. Vaniman, Luna 24 ferrobasalts and the mare basalt suite: Comparative chemistry, mineralogy, and petrology, in Mare Crisium: The View from Luna 24, R.B. Merrill and J.J. Papike (eds.), Pergamon Press, N.Y., 371-401, 1978.

Parmentier, E.M. and P.C. Hess, Spatial and temporal scales of lunar differentiation: Implications for subsequent thermal and chemical evolution of the Moon, Lunar Planet. Sci., XXVI, 1107-1108, 1995.

Pieters, C.M., J.W. Head, J.M. Sunshine, E.M. Fischer, S.L. Murchie, M.J.S. Belton, A.S. McEwen, L.R. Gaddis, R. Greeley, G. Neukum, R. Jaumann, and H. Hoffmann, Crustal diversity of the Moon: Compositional analyses of Galileo Solid State Imaging data, J. Geophys. Res., 98, 17,127-17,148, 1993.

Pieters, C.M., S. Tompkins, J.W. Head, P.C. Hess,Mineralogy of the mafic anomaly in the South Pole-Aitken Basin: Implications for excavation of the lunar mantle, J. Geophys. Res., 24, 1903-1906, 1997.

Pohn, H. and R.L. Wildey, A photoelectric-photographic study of the normal albedo of the Moon, U.S. Geol. Survey Prof. Paper 599-E, 1-20, 1970.

Ringwood, A.E. and E. Essene, Petrogenesis of Apollo 11 basalts, internal constitution and origin of the Moon, Proc. Apollo 11 Lunar Sci. Conf., Geochim Cosmochim Acta, suppl. 1, v.1, 769-799, 1970.

Scott, D.H., Geologic Map of the Eudoxus Quadrangle of the Moon, scale 1:1,000,000, Map I-705, USGS, Department of the Interior, Washington, DC, 1972a.

Scott, D.H., Geologic Map of the Maurolycus Quadrangle of the Moon, scale 1:1,000,000, Map I-695, USGS, Department of the Interior, Washington, DC, 1972b.

Scott, D.H., J.F. McCauley, and M.N. West, Geologic Map of the West Side of the Moon, scale 1:5,000,000, Map I-10348, USGS, Department of the Interior, Washington, DC, 1977.

Schultz, P.H. and P.D. Spudis, Evidence for ancient mare volcanism, Proc. Lunar Planet. Sci. Conf., 10th, 2899-2918, 1979.

Schultz, P.H. and P.D. Spudis, Beginning and end of lunar mare volcanism, Nature, 302, 233-236, 1983.

Schultz, P.H., D. Orphal, B. Miller, W.F. Borden, and S.A. Larson, Multi-ring basin formation: Possible clues from impact cratering calculations, in Multi-ring Basins, Proc. Lunar Planet. Sci., 12A, P.H. Schultz and R.B. Merrill (eds.), 181-195, 1981.

Shirley, D.N., A partially molten magma ocean model, J. Geophys. Res., 88, A519-527, 1983.

Shoemaker, E.M., M.S. Robinson, and E.M. Eliason, The south pole region of the Moon as seen by Clementine, Science, 266, 1851-1854, 1994.

Solomon, S.C., Differentiation of crusts and cores of the terrestrial planets: Lessons for the early Earth? Precambrian Res. , 10, 177-194, 1980.

Solomon, S.C. and J. Chaiken, Thermal expansion and thermal stress in the Moon and terrestrial planets: Clues to early thermal history, Proc. Lunar Sci. Conf. 7th, 3299-3243, 1976.

Solomon, S.C., and J.W. Head, Vertical movement in mare basins: Relation to mare emplacement, basin tectonics, and lunar thermal history, J. Geophys. Res., 84, 1667-1682, 1979.

Solomon, S.C., and J.W. Head, Lunar mascon basins: Lava filling, tectonics, and evolution of the lithosphere, Rev. Geophys. and Space Phys., 18, 107-141, 1980.

Solomon, S.C. and J. Longhi, Magma oceanography: 1. Thermal evolution, Proc. Lunar Sci. Conf. 8th, 583-599, 1977.

Solomon, S.C., R.J. Ahrens, P.M. Cassen, A.T. Hsui, J.W. Minear, R.T. Reynolds, N.H. Sleep, D.E. Strangway, and D.L. Turcotte, Thermal histories of the terrestrial planets, Basaltic Volcanism on the Terrestrial Planets, Chap. 9, Pergamon, NY, 1129-1234, 1981.

Spera, F.J., Lunar magma transport phenomena, Geochimica et Cosmochimica Acta, 56, 2253-2265,1992.

Spudis, P.D. and P.A. Davis, Identification of regional deposits of lunar pristine rocks from orbital geochemistry data, in Workshop on Pristine Highland Rocks and the Early History of the Moon, Lunar and Planetary Institute, Houston, TX, 69-71, 1983.

Spudis, P.D, R.A. Reisse, and J.J. Gillis, Ancient multiring basins on the Moon revealed by Clementine laser Altimetry, Science, 266, 1848-1851, 1994.

Staid, M.I., C.M. Pieters, and J.W. Head, Mare Tranquillitatis: Basalt emplacement history and relation to lunar samples, J. Geophys. Res., 101, 23,213-23,228, 1996.

Stuart-Alexander, D.E., Geologic Map of the Central Farside of the Moon, scale 1:5,000,000, Map, I-1047, USGS, Department of the Interior, Washington, DC, 1978.

Taylor, L.A., J.W. Shervais, R.H. Hunter, and J.C. Laul, Ancient (4.2 AE) highlands volcanism: The gabbronorite connection? (abstract), Lunar Planet. Sci., XIV, 777-778, 1983.

Taylor, S.R. Lunar Science: A Post-Apollo View, Pergamon, New York, N.Y., 372pp, 1975.

Taylor, S.R. and P. Jakeû, The geochemical evolution of the moon, Proc. 5th Lun. Conf. , 2, 1287-1305, 1974.

Taylor, S.R., R. Rudowski, P. Muir, and A. Graham, Trace element chemistry of lunar samples from the Ocean of Storms, Proc. 2nd Lunar Sci. Conf. , v.2, 1083-1099, 1971.

USGS, Clementine Digital Image Model UVVIS 750-nm Basemap, Volume 15, Planetwide Coverage 0.5, 2.5, &12. km/pixel, USA_NASA_JPL_CL_3015, 1997.

Walker, D., J. Longhi, E. Stolper, T. Grove, J.F. Hays, Experimental petrology and the origin of titaniferous lunar basalts, Lunar Sci. V, 814-817, 1974.

Warren, P.H., The magma ocean concept and lunar evolution, Ann. Rev. Earth Planet. Sci. , 13, 201-240, 1985.

Whitford-Stark, J.L., Charting the Southern Seas: The Evolution of the Lunar Mare Australe, Proc. Lunar Planet. Sci. Conf. 10th, 2975-2994, 1979.

Whitford-Stark, J.L., The volcanotectonic evolution of Mare Frigoris, Proc. 20th Lunar Planet. Sci. Conf. , 175-185, 1990.

Wieczorek, M.A. and R.J. Philips, Potential anomalies on a sphere: Applications to the thickness of the lunar crust, J. Geophys. Res., 103, 1715-1724, 1998.

Wilhelms, D.E., The geologic history of the Moon, USGS Professional Paper 1348, United States Government Printing Office, Washington, DC, 1987.

Wilhelms, D.E. and F. El-Baz, Geologic Map of the East Side of the Moon, scale 1:5,000,000, Map I-948, USGS, Department of the Interior, Washington, DC, 1977.

Wilhelms, D.E. and J.F. McCauley, Geologic Map of the Near Side of the Moon, scale 1:5,000,000, Map I-703, USGS, Department of the Interior, Washington, DC, 1971.

Wilhelms, D.E., K.A. Howard, and H.G. Wilshire, Geologic Map of the South Side of the Moon, scale 1:5,000,000, Map I-1162, USGS, Department of the Interior, Washington, DC, 1979.

Yingst, R. A., and J.W. Head, Lunar mare deposit volumes, composition, age, and location: Implications for source areas and modes of emplacement, Lunar and Planet. Sci. Conf. , 24, 1531-1532, 1994.

Yingst, R. A. and J.W. Head, Volumes of lunar lava ponds in South Pole-Aitken and Orientale basins: Implications for eruption conditions, transport mechanisms and magma source regions, J. Geophys. Res. Planets, 102, 10,909-10,931, 1997.

Zuber, M.T., D.E. Smith, F.G. Lemoine, and G.A. Neumann, The shape and internal structure of the Moon from the Clementine mission, Science, 266, 1839-1843, 1994.



Tables

Table 1. Listing of known cryptomafic regions identified from a literature survey. Locations of regions is illustrated in Figure 1. Areas, thicknesses, volumes, and age ranges were roughly estimated for each cryptomafic deposit on the basis of previously published literature and data (Figures 2 and 6), as discussed in the text. Area and volume totals are presented at the bottom of the table.
Cryptomafic DepositArea
(x 105 km2)		Thickness
(m)		Volume
(x 105 km3)		Age
(b.y.)
Schiller-Schickard
3.8
400
1.5
3.8 - 3.92
Humorum
2.3
450
1.0
3.8 - 3.92
Procellarum
0.38
1000
0.38
3.8 - 3.92
Australe
6.4
500
3.2
3.8 - 3.92
3.92 - 4.0
Balmer
1.7
400
0.68
3.87 - 4.1
Frigoris
3.7
400
1.5
3.8 - 3.92
Langemak
4.0
400
1.6
3.92 - 4.1
Smythii
2.0
400
0.80
3.8 - 3.84
Taruntius
0.87
600
0.52
~3.8
South Pole-Aitken
2.5
400
1.0
3.63 - 3.9
3.9 - 4.1
Schršdinger
0.05
200
0.01
3.1 - 3.87
Mendel-Rydberg
0.31
400
0.12
3.8 - 3.92
Marginis
0.29
0.67
400
400
0.12
0.27
3.8 - 4.1
3.87 - 4.1
Lomonosov-Flemming
0.52
400
0.21
3.8 - 4.0
Hercules
0.62
400
0.25
>3.2
Tsiolkovskiy
0.50
400
0.20
3.2 - 4.0
Korolev
0.95
400
0.38
3.4 - 4.04
Humboldtianum
0.21
1.12
400
400
0.084
0.45
3.8 - 3.87
3.87 - 3.92
Maurolycus
1.6
400
0.64
3.87 - 3.92
Milne
0.92
500
0.46
3.92 - 4.0
Mendeleev
0.53
400
0.21
3.2 - 3.9
Totals
35.9
 
15.6
 


Table 2. Listing of the smallest (minimum) and largest (maximum) dark-haloed craters, and their diameters (Dr), in each 100-km distance interval from Orientale, for the Schiller-Schickard cryptomafic deposit. Minimum craters are used to estimate the thickness of the overlying ejecta (tm), using equations from Chapter 2. Maximum craters are used to estimate the depth to the bottom of the cryptomafic deposit (de) from equations in Chapter 1. The thickness of the cryptomafic deposit is estimated by subtracting the minimum tm value from the depths of excavation.
Distance
from
Orientale
(km)		Minimum Dark-Haloed CratersMaximum Dark-Haloed CratersThickness of
Cryptomafic
Material
(m)
Diameter
Dr (km)		Ejecta Thickness
tm (m)		Diameter
Dr (km)		Excavation Depth
de (m)
1100-1200
4.8
317
14.4
1210
1152
1200-1300
2.9
145
21.4
2027
1942
1300-1400
3.9
257
8.5
714
629
1400-1500
2.1
139
9.4
1226
1131
1500-1600
2.7
178
9.8
823
738
1600-1700
2.9
191
10.3
865
780
1700-1800
2.2
145
15.5
1302
1217
1800-1900
4.3
284
13.4
1126
1041


Table 3. Listing of the smallest (minimum) and largest (maximum) dark-haloed craters, and their diameters (Dr), in each 100-km distance interval from Orientale, for the Humorum cryptomafic region. Minimum craters are used to estimate the thickness of the overlying ejecta (tm), and maximum craters are used to estimate the depth to the bottom of the cryptomafic deposit (th), from equations in Chapter 1. Thickness of the cryptomafic deposits are estimated by subtracting the minimum tm value from the values of th. In this region, depth to the base of the cryptomafic deposit is estimated by th because evidence has been presented (Chapter 1) that the base of the cryptomafic deposits is being reached by the largest dark-haloed craters.
Distance
from
Orientale
(km)		Minimum Dark-Haloed CratersMaximum Dark-Haloed CratersThickness of
Cryptomafic
Material
(m)
Diameter
Dr (km)		Ejecta Thickness
tm (m)		Diameter
Dr (km)		Excavation Depth
th (m)
700-800
4.3
284
9.7
514
394
800-900
9.5
627
21.3
1279
1159
900-1000
2.6
172
11.5
610
490
1000-1100
2.3
152
7.6
403
283
1100-1200
3.1
205
5.9
313
193
1200-1300
1.8
119
13.5
716
596
1300-1400
3.0
198
10.0
530
410
1400-1500
3.1
205
8.1
429
390
1500-1600
2.5
165
3.2
170
50


Table 4. Listing of all dark-haloed craters in the Procellarum cryptomafic region. Crater diameters (Dr) are used to estimate excavation depths (de) and the thickness of the overlying ejecta (tm). Thickness of the cryptomafic patches in this area is estimated by subtracting the minimum tm from the excavation depths.
Distance from
Orientale
(km)		 Diameter
Dr (km)		 Excavation
Depth
de (m)		 Ejecta Thickness
tm (m)		 Cryptomafic
Deposit Thickness
(m)
854
10.2
857
673
697
854
2.4
202
160
42
861
13.9
1168
917
1008
936
13.6
1142
898
982
946
7.5
630
495
470
1030
12.5
1050
1825
890
1066
8.5
714
561
554
1051
21.8
2060
1439
1900
1052
6.9
580
455
420




 Figures

Moon
Figure 1. Airbrush map of the Moon showing the locations of known cryptomafic regions identified in a literature survey and listed in Table 1. The lunar nearside is presented in a) , the farside in b) .



DHS'c
Figure 2. Locations and distribution of dark-haloed craters, from Schultz and Spudis [1979, 1983]. Cryptomafic deposits are often identified by the presence of dark-haloed craters. The distribution of dark-haloed craters on the Moon can, therefore, be used to give rough estimates of the boundaries of cryptomafic deposits. Note that many dark-haloed craters may not be included in this map.



Light Plains
Figure 3. Location and distribution of known lunar light plains deposits, from Howard et al. [1974]. Light plains units tend to be related to cryptomafic deposits, often obscuring them. Therefore, when light plains deposits are associated with cryptomafic materials, they can be used to give rough estimates of the boundaries of these cryptomafic units. Note that cryptomafic materials may also occur outside of light plain units.



Topography Crustal Thickness
Figure 4. Global topography a) and crustal thickness b) maps of the Moon, obtained from Clementine gravity data [Zuber et al., 1994; Neumann et al., 1996]. Mare deposits are preferentially located in basins, corresponding to areas of crustal thinning. For this reason, cryptomafic deposits are also expected to occur preferentially in basins. Thus, topography and crustal thickness anomalies can be used to help identify and define cryptomafic units.



FeO Map
Figure 5. FeO abundance map of the Moon, compiled using Clementine UVVIS data [Lucey et al., 1995, 1998a; Blewett et al., 1997]. Anomalously high values of FeO can be indicative of cryptomafic deposits and can help define cryptomafic boundaries. Only thinly covered cryptomafic deposits are expected to have FeO anomalies, thus absence of high FeO values is not necessarily indicative of a lack of cryptomafic materials.



Albedo
Figure 6. Albedo map of the Moon [USGS, 1997]. Light plains and cryptomafic deposits are both associated with albedo signatures that are intermediate between those of highland and maria. Thus, albedo anomalies can be used to delimit the extent of both light plains and cryptomafic units. It should be noted that young, bright craters may significantly raise the albedo of surrounding areas, masking cryptomafic-related albedo anomalies.



Thermal Model
Figure 7. Illustration of the lunar thermal evolution model of Solomon and Head [1979, 1980]. Partial melting through time is modeled on the basis of global and local stress constraints, where tectonic features such as mare ridges and linear rilles constrain the conversion from compressive to extensive stresses. Thermal expansion of the Moon is confined to the 1st billion years of lunar history. Thus, the heat peak must occur prior to 3.6 b.y. ago and the peak volcanic flux must also occur in this time frame.



Go to Appendix 1
Return to Chapter 4 - Table of Contents
               Chapter 3
               Thesis Title Page