TERATOLOGY IN MODERN BOTANY. 97 



The two galls differ both in form and in histological 

 structure. Does this arise from the fact that the substance 

 secreted by the parthenogenetic eggs (which substance 

 determines the gall-formation) is a different one from that 

 formed by the impregnated eggs, or is it due to a difference 

 in reaction between the vegetative point and the leaves ? 

 The reply to this question is, in my judgment, of far greater 

 importance than the investigation of apical cells and other 

 minute histological details on which so much labour is being 

 expended. 



We have already referred to the admirable adaptation 

 of the galls to the requirements of the larvte, and a point of 

 special interest is that the animal is also protected from 

 enemies during its development, partly by mechanical 

 means, partly by chemical ones (especially the abundant 

 formation of tannin), though the fact that this protection is 

 by no means absolute is shown by the frequent attacks to 

 which the galls are subjected. 



The protection required is in other cases effected as the re- 

 sult of the irritation caused by the larva in the shape of the 

 formation of roots which overgrow the larval chamber, and 

 by becoming fused together forming a close living mantle. 

 This is the case with the gall which Cecidoinyia Pocb causes 

 upon Poa nenwralis} The roots in this case arise at points 

 at which they normally are never found. Here the irrita- 

 tion which leads to the formation of the gall proceeds from 

 the larva, which attaches itself firmly to the surface of a 

 still-growing internode; fold-like swellings first appear at a 

 considerable distance on either side of the larva, and the 

 roots arise from these swellings, and may be made to de- 

 velop into normal roots if the gall be used as a slip or 

 cutting. The behaviour of Nemahis caprece however shows 

 that the gall is not always formed entij^ely as a consequence 

 of the action of the larva. The old theory of Lacaze- 

 Duthiers here applies, the development of the gall depend- 

 ing upon the substance which is introduced into the young 

 leaf with the &^'g from the poison bag. This is clear from 



^ Beyerinck, Bot. Zeitg., p. 305, 1885. 

 7 



