OF PLANORBIS AT STEINHEIM, 59 



Starting then with PI. levis, we can trace this series through a variety found in the Pits 

 which we call the ""("^g" (equal PI. m. parvus Hilg. in part), into the more angulated varieties 

 of PI. minutus. This last figured on pi. 9, fig. 17, and line e, fig. 16, pi. 2, is the PI. 

 '^iSltut^a of Hilg. Notwithstanding this name it will be found by comparison with such 

 forms as have just been mentioned, as PI. "Ss"; ^^bat it differs only in size. 

 This leads into a variety of PI. minutus, fig. 18, pi. 9, and fig. 1, line e, pi. 2, in 

 which the breadth and angularity of the inner part of the aperture is still maintained, but 

 the shell is otherwise a variety of PI. minutus, and has the thinner proportions of the 

 young whorls observable in this species. Both this and fig. 17, pi. 9 seem to fade by 

 imperceptible gradations into fig. 19, PI. triquetrus, also figured in nos. 1-3, pi. 4, line c. 

 Or they may be traced into the less involute forms of PZ. minutus. These have exceedingly 

 cylindrical whorls, and umbilici entirely open on both sides, and connect the more involute 

 or levis-like m,inutus forms, with the uncoiled PI. denudatus. 



Rem,arks: On pi. 4 this genetic series is fully illustrated. Line a exhibits various va- 

 rieties of PI. viinutus. These appear to be identical with the PI. m. minutus, var. ^ 

 teres of Hilgendorf in part, and in part are equal to his PI. m,. minutus, var. a tyjjus, 



I think that figs. 1, 3, 5, 7, line a would be considered as belonging to the first named. 

 The figures on line a, however, are arranged with the view of exhibiting the varieties 

 which tend to deflect the whorl by growth against the spiral. Fig. 7 is an extreme form 

 in this respect. Figs. 2, 4, line &, are typical minutus forms, and connect directly with 

 PI. 'm?M«S' figs. 7-8, line 6. These have a closer umbilicus on the lower side (compare figs. 

 2 and 6, line 6), and in this respect approximate to PI. triquetrus. This last named species 

 is figured on line c, figs. 1-5. PI. triquetrus, var. turbinatus figs. 5 and 9, line b, the latter 

 a section, exhibit very distinctly the tendency towards trochiform growth of the spiral, 

 which is common in all the species and varieties at Steinheim. This also, is the proper 

 place to notice the modes of variation among Steinheim shells. 



It will be observed that the varieties of any one of the forms previously de- 

 scribed fluctuate between two extremes so far as the spiral is concerned. As in the case 

 •before us these extremes are indicated by a tendency to reverse the spiral in some forms, 

 and to increase it in others. This correllates with a widening of the mnbilicus on the 

 lower side in the former, and a narrowing of the umbilicus in the latter. 



These two again correllate with more cylindrical whorls in the former, and wider or 

 larger whorls in the latter. The latter correllation is important)^ since it enables us to 

 draw one more important distinction between the healthy and unhealthy series, or the 

 progressive and retrogressive, as 1 have called them. Thus on the Summary Plate, pi. 9, 

 series 2-4, exhibit this in a marked manner, and so also does sub-series 3, which thus 

 shows another progressive characteristic besides those previously enumerated. 



If there is any truth in the assumption that health marks the favorable character 

 of the surroundings, and that such correllations are signs of healthy growth, then 

 the sub-series previously assumed upon other grounds to be diseased or unhealthy, 

 might show the universal tendency to form a spiral, but ought to be deficient in 

 healthy characteristics. I have already shown this to be the case in different degrees 

 according to the character of the sub-series, and I now have to add, that they are in a measure 



