140 is some measure of the imperfection of the geological record. In 

 order to formulate hypotheses concerning the history of the fauna in 

 this study, the actual known occurrences have been taken at their 

 face value ; the earliest record of a genus is inferred to be its first 

 appearance in New Zealand and the absence of a genus after a period of 

 occurrence is taken to indicate its extinction. The arbitrary acceptance 

 of negative evidence implied by this empirical method is unavoidable 

 if we are to generalize at all, but is a warning that such generalization 

 is fallible and liable to revision with improvement of the data on which 

 it is based. 



Faunal Richness 



The total number of genera present in any one stage or substage 

 varies, and may be used as an index of faunal richness (Fig. 1). In the 

 Pliocene, 452 genera occur, and the total present at one time varies 

 from 327 to 354. The early Pliocene faunas (Opoitian and Waipipian) 

 are richer than later ones, and there are two periods of relatively low 

 totals in the Mangapanian-Hautawan and in the Okehuan. Fluctuations 

 in faunal richness are the result of extinctions on the one hand and 

 invasions on the other. The form of the graph suggests that these 

 opposing processes did not operate at random, but that the phases when 

 extinction was the dominant process alternated with phases of invasion. 



350 



340 



330 



WM WW O 



Fig. 1. 



-Total number of molluscan genera in New Zealand Pliocene and Pleistocene 

 stages, plotted against time scale. For symbols see Table 1 



Numbers of Immigrant Genera and of Extinctions 



Figure 2 summarizes what is known of the appearance and dis- 

 appearance of molluscan genera during post-Miocene time in New 

 Zealand. The total " first appearances " for a given stage or substage 



310 



