OF PLANORBIS AT STEINHEIM. H 



Figs. 5-10 and 6-11-20, at first sight do not seem to be very close ; but if we 

 remember that PI. discoideiis, fig. 5, and PI. "'cul'mdms^ fig- 6, often have no carinations on 

 the lower side, and if we compare these and such forms as are figured on pi. 2, line c, fig. 

 1-6, with the sulcated forms of PI. stqjremus, the resemblances will be seen to be quite as 

 close as in other representative forms. Here also thei'e can be no doubt of the absence of 

 hybrids. 



PI. trochiformis, fig. 7, stands alone in spite of its general resemblance to clenuda- 

 tus, fig. 24, because the similarities to the latter are not exhibited in the form and 

 characteristics of the whorl itself, but are simply such as any spiral shell might have 

 to any other allied form, with a similar spiral mode of growth. 



This tendency reaches a climax in both the second sub-series and the Fourth Series, and 

 thus, though PL trochiformis and denudatus (trochiform variety, pi. 9, fig. 24), have 

 very distinct whorls, the general outline of the whole shell in each is similar. This is 

 especially the case when we compare an old or distorted PL trochiformis, pi. 2, line r, 

 fig. 10, in which the last whorl is j^^rtly unwound, with cZemtrfo^i/s, in which the young 

 whorls are also closely wound, and the last whorls only open. This peculiarity has a dis- 

 tinct meaning from that of the representative characteristics described above, and will 

 be discussed farther on. 



At present it will be better to concentrate attention upon the representative charac- 

 teristics in the progressive series, (Second, Third, and Fourth Series). As has been 

 said above, we can readily account for the production of representative forms on the lower 

 lines, because these are really one and all but slightly different from the four original varie- 

 ties of PL levis, and resemble each other necessarily on account of their close affinity, 

 though not joined by hybrids. But how can we account for the fact that still more 

 divergent forms, which follow and which have less affinity with PL levis, should resemble 

 each other, species for species, with such remarkable closeness ? New modifications are 

 introduced in these species, which are not present in the original species. These new 

 characteristics consist of altei'ations in the forms of the whorls, and the advent of sulca- 

 tions, accompanied by a more decidedly asymetrical mode of growth. These modifications 

 could not have been inherited from PL levis, since they did not exist in that species, nor 

 in the lower representative foi'ms, pi. 9, figs. 2-9-18. 



These considerations enable us to separate the representation in the parallel series into 

 two kinds, that which occurs by the inheritance of a similar form and characteristics from 

 the parent form, PL levis, and that which results from the introduction of new 

 characteristics in each series, which are similar to those produced in corresponding forms 

 in other series, and which could not have resulted from intercrossing of the different 

 species in the separate series. 



Two of the progressive series, the Second and Third, present us no foi'ms strictly 

 parallel or representative to PL tenuis, fig. 40. This can be readily explained by the 

 fact that these series retain with great pertinacity the characteristics of the varieties of 

 PL levis, from which they sprang. The Second Series remains smooth and devoid of 

 sulcations, though the outer side of the whorl becomes angulated in PL crescens. The 

 Third Series retains the gibbous form of the under side of the whorl, and does not flatten 

 the whorls at all, as in PL tenuis. Nevertheless, the general tendency to increase the 



