174 



PSYCHE. 



[February — March 1SS9. 



that these protective adaptations are the 

 result of natural selection, acting through 

 the ages of co-existence of plants and 

 their enemies. I have long felt a con- 

 viction that the occurrence of extra- 

 nuptial nectar-glands on so many of our 

 own plants which are not menaced by 

 leaf- cutting ants (except in the extreme 

 southwest) , must date back to later 

 geological periods, characterized by a 

 warmer climate in northern latitudes, 

 and by a much greater prevalence of 

 ants of many kinds than is the case with 

 us now* ; and that a corresponding prev- 

 alence of leaf-cutting or other noxious 

 ants is demonstrable in these periods. 

 Unfortunately, I have never been able 

 to pursue the subject in this direction, 

 but the immense collection of fossil ants 

 in the possession of Mr. Scudder must 

 furnish instructive data for testing this 

 opinion, when its treasures shall have 

 been sufficiently studied to show the 

 affinities of the prevalent genera of those 

 times in northern America. That struc- 

 tures corresponding in position to the 

 foliar glands of existent species of Popu- 

 lus were found on tertiary plants, is 

 shown by their occurrence on P. glan- 

 dtcUfera., Heer (31, p. 390), but I do 

 not now remember that they have been 

 recognized elsewhere. 



While, as has been shown, the secre- 

 tion of food for protective ants is often 

 rendered superfluous by the provision of 

 other more direct deterrents, the two 

 classes of protective adaptation sometimes 

 occur conjointly. Lundstrom (16, p. 



* On this point see a very instructive note by Mr. 

 Belt in Nature, 1S77, v. 16, 122. 



83) has shown that the leaves of Popii- 

 lus which do not possess nectar-glands 

 are distinguished also by the thinness 

 and flexibility of their compressed peti- 

 oles, by which caterpillars, etc. are to 

 some extent rendered uncertain in their 

 footholds on these easily shaken leaves. 

 Delpino and Schimper have also ob- 

 served the protection secured to Ricinus 

 by its smooth and very glaucous stems, 

 though its leaves likewise bear extra- 

 nuptial glands. 



An aberrant function of the nectar- 

 glands on the leaves of some pitcher 

 plants is that of luring to their destruc- 

 tion, insects which are captured and di- 

 gested or at least macerated by these 

 carnivorous plants. From the large 

 number of ants fovmd in our Sarracenla 

 leaves in a state of nature, this would 

 seem to be true of species of this group ; 

 for these ubiqviitous insects are certainly 

 led to the orifices of the pitchers by the 

 sugary secretion on the exterior. In his 

 latest paper (8, 227), Delpino holds this 

 secretion to be protective, as in the 

 cases already passed in review ; but the 

 opinion which I have here and elsewhere 

 (30, 328) expressed, is that of Riley (25, 

 p. 25), Mellichamp (18, 119), Gray 

 (10, 112), and others, some of them 

 early writers on Sarraceniaceae. That 

 ants are largely victimized by these 

 plants does not, of course, signify that 

 the structure of the latter is not such as 

 to facilitate the capture of larger, flying 

 insects, which are, in fact, often found 

 entrapped, especially by the southern 

 Sarracenias and the Californian Dar- 

 lingtonia. 



