458 



SCIENCE. 



[N. S. Vol. I. Ko. 17. 



the older American species, but it is very 

 imperfect in taking no note of the many 

 later described American species, especially 

 those described by Ashmead and Gillette. 

 The classification is based on Mayr's, as was 

 that given in Lichtenstein's translation of 

 1881, and comparatively few additional spe- 

 cies are included. 



The introduction is verj^ full and includes 

 a discussion of heredity and a rather full 

 summary of late embryologic work, with a 

 view of getting a clearer conception of the 

 philosophy of alternation in generations. 

 Mr. Straton particularly discusses Weis- 

 mann's views, but by no means accepts 

 them, though a thorough believer himself 

 in natural selection. 



Straton points out " that galls may be ar- 

 ranged in groups of greatly increasing com- 

 plexity and that they must have arisen by 

 gradual and complete improvements in the 

 initial stages of their formation, acting 

 through natural selection over an unlimited 

 period of time and through numerous con- 

 secutive species." Each infinitesimal im- 

 provement in the gall itself, internally or 

 externally, which has been of service as a 

 protection against parasites or as favoring 

 the development of the larva, has been pre- 

 served. In this view of the case, which is 

 one that certainly seems most reasonable, 

 the various characteristics of galls, such as 

 spines, prickles, glutinous secretions, indui-- 

 ation, and even size and coloration, are all 

 acquired characteristics for the protection 

 of the larva within. This theorj^ is cer. 

 tainly justified in a large number of cases, 

 but is equally at fault in manj^ others. It 

 would be hard to conceive that the bright 

 colors which many galls assume in an early 

 stage of development or the succulent char- 

 acter and pleasantly sub-acid or fruity 

 flavor of others which renders them so prone 

 to be invaded and preyed upon hj a host of 

 other insects could have any relation to the 

 benefits of the gall-maker within. Here, as 



in most other natural history phenomena, 

 natural selection can hardlj^ be considered 

 an all-sufficient explanation. Likewise, the 

 assumed protective colors which galls often 

 take on in autumn will find more valid ex- 

 planation in the same causes which produce 

 the similar changes in the leaves themselves, 

 which can have no reference to the welfare 

 of the plant. 



No subject connected with galls has per- 

 haps been more written about than the in- 

 citing cause of their formation. Adler and 

 Bj'erinck effectuallj' disproved the older be- 

 lief that the exciting poison was inserted by 

 the parent in the act of oviposition, i. e., 

 that the initial force was due either to a 

 chemical secretion injected by the gall- 

 mother or to the mechanical stimulus of 

 traumatic ii-ritation. A fluid is secreted in 

 the act of oviposition, but it is absolutely 

 unirritating and acts primarilj- as a lubri- 

 cant to facilitate the arduous mechanical 

 act and probablj' also as a mild antiseptic, 

 dressing to the wound made in the plant. 

 Nevertheless there is an irritating salivary 

 secretion produced by the larva itself and 

 the gall gTO\vth is co-incident with the 

 hatching and feeding of this larva. The 

 fact that the influence on the plant tissues 

 sometimes begins before the egg-shell is 

 ruptured indicates that this fluid possesses 

 amj'lolytic and proteolj'tic ferments. That 

 the influence should be slightly exerted pre- 

 natallj' is not to be wondered at when we 

 consider the delicate nature of the egg cover- 

 ing which often makes it diflicult to observe 

 the dividing line between the egg and newly 

 hatched lai-va. 



While, therefore, it is the larva in the 

 Cynipidfe which causes the gall, this is not 

 the case with the manj' other gall-produ- 

 cing insects, since manj' of the gall-gnats 

 (Cecidomyidte) and most, if not all, of the 

 gall-making saw-flies (Tenthredinidse) se- 

 crete a poison in the plant tissue in the act 

 of oviposition, causing the gall to form be- 



