OF PLANORBIS AT STEINHEIM. 



65 



tent inherited tendency to fonn a closely coiled shell during the protected stages of the 

 earliest period of growth. I failed to find any finely costate forms, with distorted or 

 open whorls. 



Line h, pi. 4, gives photographs of the variety major of Hilgendorf, and he probably 

 also includes in this variety the coarsely costate forms on line k. 



Var. major {PI. multiforinis var. major Hilg.), fig. 15, line h, pi. 3, is an exceedingly fine 

 specimen of the largest size. It is comparatively rare even in formation n of the East 

 Pit, where it was most abundant. It was, however, not difficult to ascertain that it was 

 divisible into sub-varieties, having fine and coarse costae, but these are invariably more 

 or less widely separated, and therefore belong to the coarsely costate series. 



The following table depicts these relations diagramatically for the purpose of placing 

 them in a clearer light. 



Coarsely costate sub-series. 



PI. costatus var. 



platystomus. 



PI. costatus v.ar. 

 obtuso-costatus. 



Finely costate suh-series. 

 PI. costatus var. 

 distortus. 



PI. costatus var 

 acuto-costatus. 



Smooth shelled sub-series. 

 PI. denudatus. 



PL minutus. 



PI. 



costatus 

 minutus. 



PI, 



costatus 

 minutus. 



PI. minutus. 



-PI. minutus.- 



Pl. minutus and its immediate affinities are shown by three sub-series. Each of 

 these have the cylindrical and less involute forms corresponding to PL minutus, and 

 the completely trochiform and partly uncoiled cylindrical whorled forms, the equivalent 

 of PI. denudatus. Thus each of the three sub-series presents a similar succession of forms, 

 the ancestral, or closely allied, the highly differentiated or distinct forms of the third line 

 and the diseased and closely representative forms of the fourth line. 



I do not think that the accepted limitation of a species to one or more series of forms 

 connected by hybrids or intermediate varieties, is of any use whatever in estimating 

 the value of the characteristics in cases like that under consideration. The value 

 of these must be determined with reference to all the members of the group in which 

 they occur ; this alone can give their approximate taxonomic meaning. Thus, by 

 reference to the Planorbidae generally, we can show that the modification represented 

 by forms in the smooth and costate series, are really more distinct than most of the 

 species of the genus Planorbis. If the intermediate forms were lost or destroyed, 

 there would be no doubt on this point. If the word species can be used to mean 

 anything at all, it must be restricted in given groups to certain limited series of 

 modifications, having a certain approximately determinable value. If the term can 

 be used at one time to designate so great a series as is included from minutus to 

 costatus, or Steinheimensis to trochiformis, or even discoideics to trochiformis, I can 

 see no reason why at another time it may not be used for all these forms together, as 

 Hilgendorf has done. 



Fig. 15, line b, pi. S, has no costae on the last part of the outer whorl, and this repre- 

 sents the extreme old age condition of the costate series. This return of the smooth 



