Personal experience in Pendejo Cave and in the laboratory supporting the investigations there bears directly on claims for stratigraphic integrity forwarded by the authors of a report on ancient friction skin imprints. The published report is seriously incongruent with my notes and recollections. The statement that my work indicated no intrusions directly above the first recovery of imprints contradicts the fact that I spent my excavation time there clearing a large rodent burrow that penetrated to the layer in which the find was made. Abundant cultural remains of Holocene age in the uppermost strata of the cave constitute potential sources for anthropogenic items of all kinds that could have been intruded to great depths by burrowing fauna. Data supporting dismissal of such possibilities must be presented by the investigators on a firmer basis than given in their 1996 report.

Only in the last few years have high-resolution paleoclimatic data become available from coastal southern California. Recent research in the California Channel Islands, drawing on some of these data, attributes settlement disruptions, disease, and violence to maritime subsistence distress attendant to elevated sea temperatures in the period from A.D. 1150 to 1300. A broad range of paleoenvironmental, archaeological, and human osteological data suggest that these stress indicators are more convincingly correlated with severe late Holocene drought episodes during a portion of the medieval climatic anomaly (ca. A.D. 800 to 1400). Based on these data, cultural changes in coastal southern California, including violence, declining health, and emergent social complexity, are similar to events documented in the American Southwest. Cultural adaptations in both regions appear to have been responding to persistent drought conditions during the late Holocene.

We are in full agreement with Leonard, and many others, that the sample size-richness relation is an important concern for archaeologists, and we appreciate his evenhanded discussion of our paper. However, we disagree with Leonard regarding the point in the research process when variable prehistoric behavior can be used to explain differences in richness. Leonard and others argue that we must always begin studies of richness by controlling any observed relation to sample size; we can only consider prehistoric behavior as an explanation of richness after we account for sample size effects. In contrast, we suggest that following his recommendations will obscure aspects of behavioral variation that are important to understanding prehistoric societies. Considerations of prehistoric behavior must therefore be central in studies of richness.

Eubanks bases her identification of the fossil pollen grain from Alabama as Tripsacum primarily on her calculated spinule density. To make those calculations, she used only our published photograph, and she assumed a grain expansion of 35 percent. She ignores the fact that the spinule density of the fossil pollen grain is actually the same as that of similarly treated Zea mays pollen. While there is always the possibility of a misidentification or of long-distance transport, the most likely interpretation remains that the 3500 B.P. pollen grain is Zea mays and that it represents limited cultivation of ancient corn in southern Alabama.

Although the relationship between settled area and population size has been well studied in archaeology, anthropology, and geography, with numerous empirical data sets, an established methodological approach, and a body of theory, Schreiber and Kintigh's (1996) analysis of archaeological and historical data from the Peruvian Andes found “only a weak correlation” between these variables. By employing double logarithmic data, the relationship is shown to be much better than the “surprisingly poor” one the raw data suggests. Regression analysis of these Peruvian data in logarithmic form yields power functions that conform closely with expectations derived from a body of settlement size theory based on the allometric principle.

This paper corrects a number of false assumptions made by Dincauze, among them the following: (1) her work area was in the southern half of the unit, not the undisturbed northeast quadrant below which imprints were found; (2) possible rodent activity was reported on profile drawings (19) and in excavators' notes and floor plots (121 instances); (3) no endorsement from Dincauze was expected or desired; (4) the Paleoindian claims are valid by Griffin's criteria; (5) MacNeish has followed high standards of reporting for decades and will continue to do so.

While new data about drought cycles are informative, patterns of ill health and violence and the origins of sociopolitical complexity in southern California are best understood through a broad consideration of multiple conditions and processes. These include various kinds of environmental and demographic conditions, changes in social and economic strategies, and changes in the organization of labor.

The concern with the gender and role of Mimbres potters and painters is certainly an important one, and Hegmon and Trevathan (1996) do a good job of reviewing the various arguments about the identity of the bowl painters. Going beyond this review, they add another possible piece of evidence for the role of men in the production of Mimbres art. While I concur that the other arguments they present suggest some role for men in the painting activity, I do not believe that the particular new example they present adds any weight to this argument.

The identification of human friction prints by Chrisman et al. (1996) from Pendejo Cave, New Mexico, some dating more than 36,000 years B.P., is tenuous given the incongruence between data presented in the text and data illustrated in the figures. The identification of these structures as human friction prints hinged largely on the frequency of epidermal ridges per linear centimeter. As such, the validity of this evidence to justify the presence of humans at Pendejo Cave at this early date is questioned.

In a 1995 paper, Raab et al. questioned Arnold’s assertion that high sea temperatures in prehistoric southern California resulted in maritime subsistence disaster. The current discussion by Arnold et al. argues that our data were flawed and taken out of sociocultural context. Not only are these claims unfounded, they are based largely on an unfortunate strategy of combining uniformitarian and interpretivist analytical frameworks

Accurate measurement of the number of epidermal ridges must be done directly from the artifacts, using a nearly flat surface or a very thin dental silicone mold pulled off and flattened. Measurements from enlarged pictures of 3-D rounded objects converted to 2-D are highly inaccurate and do not correct for the nearly 20 percent shrinkage factor.

Based on submission forms and other documents deposited in the Smithsonian Institution archives on termination of the Smithsonian Radiocarbon Dating Laboratory in 1986, Anna Roosevelt argues that shell middens on the coast of Guyana and northeastern Brazil contain pottery, and that the dates support her argument that “Amazonian early pottery is the most securely dated early pottery in the New World” (1995:128). Pending publication of a detailed monograph, I maintain that the Guyana sites in question are preceramic and thus offer no support to Roosevelt's thesis.

Hegmon and Trevathan's (1996) contention that Mimbres figurative pottery was probably painted by men ignorant about birthing, as evidenced by a virtually impossible birthing scene, is unsubstantiated. This claim was due to a lack of familiarity with the birthing motif they described, as well as selective and unrepresentative use of relevant ethnographic and historical data. We find no evidence that men painted Mimbres pottery.

Denis Williams writes to comment on my article on Archaic shell mound pottery in eastern South America (Roosevelt 1995). He states that he will “correct” my article by putting on record “new facts.” Rather than correct my article, Williams"s comment misstates both the content of my article and that of earlier literature on Guyanese archaeology, and it merely repeats the data included in my article. In addition, Williams's comment presents some interesting but internally contradictory elaborations of his earlier interpretations of Guyanese archaeology but still without supplying the basic data on which his interpretations are based. In essence, contrary to my article, Williams states that there is no such thing as a Guyanese Archaic shell mound pottery occupation, known in earlier literature as the Alaka Incipient Ceramic phase (Evans and Meggers 1960:25-64). Williams presents this conclusion as “fact,” but it contradicts the existing data from stratigraphy, pottery distribution, and radiocarbon dates in the shell mounds, and he furnishes no other specific data that support it. In my comment on his comment, I will document these various aspects of his comment and define the type of data that he needs to present to allow empirical evaluation of his assertions.

In their 1996 article, Hegmon and Trevathan argue that the nature of birth scenes depicted on Mimbres pottery bowls provides insight into the gender of the pottery painters. Specifically, they argue that the birth position illustrated is extremely uncommon, suggesting that the potter(s) had not had exposure to birth episodes. Hegmon and Trevathan conclude that the potters were male. Their arguments are flawed by faulty premises and a severe misunderstanding of the nature of Mimbres art. A review of the evidence regarding birth-depiction bowls indicates that the images do not imply either male or female potters.