1G2 Records of the Indian Museum. [Vol. XIY, 



T. theohaldi is a fairly constant form. Shells from 3,000 feet do not differ, 

 except in sucli accidental characters as being more free from a coating 

 of calcareous deposit, from shells from 5,000 feet, nor do those from 

 ancient deposits differ from those found with the animal alive. The 

 fact that shells from Kalaw are coated in the manner indicated while 

 those from Yawnghwe are nearly clean proves in itself that the two lots 

 were living under different conditions ; that they closely resemble one 

 another proves that the shell of the species is not plastic. In some 

 at any rate of the streams in which it occurs T. theohaldi is subjected 

 to the influence of frost. It is not always easy to distinguish single 

 shells of T. theohaldi from selected shells of the other non-lacustrine 

 species (T. naticoides), though their radular teeth differ considerably 

 in proportions ; but T. naticoides, which lives in swamps, pools and 

 the back-waters of larger streams, never so far as we know at very 

 high altitudes, is an extraordinarily variable species in respect not only 

 to shell-sculpture, but also to colouration and to shell-form. Theobald 

 in his original description recognized three " varieties," which he named 

 " the typical form," var. fasciata and var. carinata, but he acknow- 

 ledged that it was not possible to draw an exact line between them. 

 The shell of his typical form (afterwards named var. concolor by 

 Nevill) was almost smooth, while those of the varieties fasciata and 

 carinata had spiral ridges developed in different degrees. So far as 

 these characters are concerned, Theobald's observations are fully borne 

 out by my own, but a large series of specimens does not support his 

 assumption that differences in colouration and shape of shell are always 

 correlated with differences in shell-sculpture. The more highly orna- 

 mented shells are often more conical than others, but beyond this 

 I cannot go. The differences that do exist are entirely individual and 

 in a single handful of shells from a single pool one may find specimens 

 that are nearly smooth, specimens in which the sculpture is highly 

 developed, and all intermediate phases. As I will show presently the 

 proportionate number of individuals that can be assigned approximately 

 to one variety or another differs in different localities. 



In describing the three varieties Theobald, probably with a limited 

 number of shells before him, gave due attention to colouration, shape 

 and sculpture. His specimens were apparently all from the Upper 

 Salween, several hundred miles from the localities in which we collected. 

 I have taken at random one hundred shells from a series collected in a 

 sluoo-ish stream on the He-Ho plain. Of these nine have no definite 

 spiral ridges, but none are absolutely smooth (see figs. 3, 4, pi. xvi). All 

 these shells without spiral ridges are distinctly banded in colouration. 

 I found fifteen shells that could be definitely assigned to the var. carinata 

 so far as shape and sculpture were concerned, but unbanded, while the 

 remaining seventy-six shells formed an unbroken series between carinata 

 and fasciata in shell-form and sculpture, but were all banded. The nine 

 banded, comparatively smooth shells were all of a somewhat globose 

 type, on an average distinctly less conical than the carinate examples, 

 but no difference of a kind that could be measured or accurately estimated 

 in any other way was observed. The average size of fully adult shells 

 in this series (length 35-38 mm.) is greater than was probably the case 



