140 TRANSACTIONS OF THE [MAR. 9, 
to seek some sort of explanation of the facts we are dealing 
with. 
If we found these successive modifications of the type form 
in successive geological strata, we should of course conclude 
that these varieties were on the way to becoming species, in- 
other words that we had to do with a case of evolution. But 
here we are dealing with contemporaneous forms inhabitating 
apparently the same habitat. Mr. D. C. Worcester, who col- 
lected these shells, informs me that he was unable to discover 
any evidence of local distribution. He states further that, with 
regard to another Cochlostyla nearly as variable as this, he can 
state positively that the various extremes are often found feed- 
ing upon the sameclump of bamboo. Quadras writes that those 
found in the valleys are generally smaller than those found near 
the seashore, but mentions no local distribution in reference to 
color. Moreover we have no evidence to show that these vari- 
ous extremes do not freely interbreed or that the most unlike of 
them may not hatch from the eggs of the same parents. In- 
deed, if there is no local distribution, it is inconceivable that the 
different varieties should breed true. 
Consider also that no theory of evolution yet offered is com- 
petent to explain how these varieties can become fixed species. 
Any theory which appeals to the direct action of environment 
has no application, for here the environment is the same for all. 
Against the operation of Natural Selection we can offer objec- 
tions equally valid. We should be forced under this hypothesis 
to assign some utility to these markings, the most plausible 
view being that they are of protective value. But Mr. Worcester 
writes, in reply to a question of mine on this point: “After 
thinking the matter over a good deal, I incline to the opinion 
that we have to do here with variations neither especially useful 
nor especially harmful.” And, indeed, it is in the highest de- 
gree improbable that animals with like habits could be protected 
equally well by such a number of colors. Again, if Natural 
Selection were here in operation, we should expect to find the 
terminal forms of the tree, z. e., the most improved, to prepon- 
derate greatly in number, for Natural Selection can only operate 
by the extinction of the unimproved. But a count of the speci- 
mens finds the reverse to be the case: 
es Kae ep a a alae heal Cd 5A ee a aa Me fs 318 
BIOS Lys cueote sags cL en ee eae, ee 41 
VATA SAL Ea ants Waren MeV tte Marin 2 SOR s re 18 
aR Blacon veka ayatasand herons vigWus de tok Medica a aioe Pe ahek an 66 
Gil sea Narn cs Osa ashw od ch et sale toe mae Reh ok an 16 
MO TREE cos ccna cg eatagh sdeaeh eae oheaeae: 67 
