144 J. Rki.tnvaax — Doch Its a'. I.kkuwex 



(iOTilrt not get till very late , the question is also pnt if the nnni- 

 brous galls of inites would really be derived from as many sorts. This 

 question is very diftieult to answer , as the elassifìcation of these 

 uiites is so hard. 



Let us consider all this from a jdiylogenetie point of view. If 

 we take a plant, that is in a period of mutation and s<> at different 

 times pro<luces new s])ecies , and this plant gets infected by a cer- 

 tain gall-insect, the insects emanating from it ccnild only infect new 

 plants, if the characters had not changed in their ai'angeuient neces- 

 sary to the gall-formation, or as could be the case too, that the in- 

 sect was kept to that same kind by its instinct. If the possibility 

 of gall-formation were to be excluded by the mutation , the insect 

 must have a mutation too to keep pace with the changes in the 

 plants. And jierhaps we c<nild l»y this come to a better understan- 

 ding of the fact, tbat of plants which differ scarcely from each other 

 one is strongly infected by galls and the other very little. Is it not 

 necessary for i)ractical purposes to cultivate jdants , which are im- 

 mune to certain parasites, and does not it happen sometimes, that, 

 what is got by accurate and careful cultivation gets all at once in- 

 fected by the same disease as raged formerly ? Can this really be 

 the same disease ? We may doul)t it and from uniny facts , which 

 are ])ublished lately , we uiay accept , that the i)arasites can only 

 follow the progress of tiie plants , when they themselves have by 

 mutation got the (;apability to form new galls on new idant-species. 

 For the tmie all this is onl>^ very imcompletely proved, but surely 

 it will be shown clearly by exact investigation of the galls and gall- 

 makers of closely related forms. Only with some species the expe- 

 riment has been undertaken and , as our experiments show, mostly 

 with negative results. 



So we want to infer : the insects, which form galls on closely rela- 

 ted species and are not morphologically to be distinguished from each 

 other, differ physiologically so as to form a number of elemental species. 



Where it has lately been clearly shown by the researches of 

 Jordan (1) on D r a b a verna and Viola tricolor etc. that 



(1) Jordan- Citated after H. dk Vriks. Die Mutationstlieorie. Leipzig- 1901 

 Bd. I. pag. 16 and 121. 



