THE SPECIES PROBLEM WITH FOSSIL ANIMALS 



Samples of the sort just described are by no means common, 

 however. Four size-frequency distributions typical of fossil ma- 

 rine invertebrates are given in Fig. 10. Such distributions tend to 

 be unimodal and either symmetrical or slightly skewed; many 

 can be approximated by the normal distribution. In most in- 

 stances, the observed distributions are strongly biased and reflect 

 primarily a complex of ecologic, geologic, and operational factors 

 which at best bear only obliquely on the taxonomic problem 

 (Boucot, 1953; Kermack, 1954; Imbrie, 1955). 



In dealing with groups of animals (mammals, for example) 

 where practical and objective morphological criteria are available 

 for the identification of definite growth stages, the existence of 

 biased size distributions may cause little concern to the taxono- 

 mist. In dealing with fossil groups lacking criteria of this sort, 

 however, particularly in groups displaying strongly allometric 

 growth patterns, taxonomic judgments may be considerably af- 

 fected. Once recognized, this problem can be quite easily solved 

 by characterizing sampled populations in terms of growth pat- 

 terns rather than growth stages ( Kermack, 1954; Parkinson, 1954; 

 Imbrie, 1956). 



3. Incompleteness of the available fossil record. The most 

 serious limitation of paleontological data is the sparsity of fossils. 

 It is of course true that the total number of collecting localities 

 which have yielded good fossils from every system younger than 

 Precambrian is very large, and it is also true that future work will 

 bring forth an unknown but certainly very large amount of new 

 material from localities and horizons now unrepresented in exist- 

 ing collections. Nevertheless, from general theoretical considera- 

 tions on the nature of sedimentation and diagenesis, and from 

 practical experience in portions of the geological column which 

 have been thoroughly examined for fossils, most paleontologists 

 and stratigraphers would predict that no amount of future field 

 work will ever fill a majority of existing phyletic gaps between 

 transient species. 



At least five factors are important in accounting lor the incom- 

 pleteness of the available fossil record, not including lack of ade- 



