in VARIATION UNIVERSAL 149 



has been the case with the common pansy, as Weis- 

 mann has noticed, and it often happens that the 

 foreign species, after keeping very constant for some 

 years, begins all of a sudden to vary considerably and 

 in many directions, thus giving birth to an unexpected 

 number of varieties. These facts must be kept in mind, 

 and they go to show that one must not be hasty in 

 deciding whether any species is or is not liable to vary 

 more or less than another : variability is doubtless 

 itself variable, according to influences of which we are 

 more or less ignorant. Palaeontological facts are 

 known which are of value in that they show that 

 variability has existed in the past as well as in present 

 times, and that it has been more apparent in some 

 species or genera than in others. Mr. Hyatt provides 

 a good instance of such facts, in his paper on the 

 Steinheim fauna, 1 where a large number of fossil 

 varieties of Planorbis are met, all of which most 

 probably descend from four principal varieties of one 

 single species, P. laevis, unless we prefer to believe that 

 every single one of these principal or secondary 

 varieties has been called into existence by an equal 

 number of special creations. H. Filhol has also given 

 valuable facts bearing on this matter, in his investiga- 



1 See The Genesis of the Tertiary Species of Planorbis at Steinheim 

 (Bost. Soc. Nat. Hist. 1880), and Transformations of Planorbis at 

 Steinheim (Am. Naturalist, 1882, p. 441.) Cf. also Stearns, Proc. 

 Acad. Nat. Sc., Philadelphia, 1881. 



