BIOLOGIC RELATIONS BETWEEN PLANTS AND ANTS. 421 



A study of the development of the leaf suffices to show that, mor- 

 phologically, these represent aborted teeth of the limbus that have 

 become adapted as nectaries. Between the serrate teeth that form tbe 

 edge of the limbus of a number of leaves there are found in various 

 plants small nectariferous glands (serration glands). In the wood- 

 vetch the stipules, situate at the base of the compound pinnate leaves, 

 serve as foliary nectaries. 



It is quite natural to ask what can have been the cause of this 

 localization of the productiou of honey dew at this particular point 

 to the exclusion of others. The following explanation is reasonably 

 satisfactory : 



Nectarivorous insects, having acquired the habit of frequenting 

 leaves covered throughout their entire surface with honey dew, con- 

 tinued to do so, even when the excretion had ceased; during periods, 

 for example, when transpiration was not retarded. These leaves they 

 subjected to suction, and if their buccal apparatus permitted it, as in 

 the case of ants, to a continually repeated nibbling. In this respect 

 these insects behaved like a young mammal who sucks the breast of 

 his mother more energetically in proportion as she furnishes less lac- 

 teal secretion. Any irritation of a living tissue causes it to hyper- 

 trophy and proliferate. The localization of the irritation at certain 

 special points causes the formation, at these points, of glands having 

 a sugary secretion. Henceforward the nectar ivorous insects localize 

 their action upon these nectaries, and the remainder of the leaf may 

 then adapt itself entirely to other functions, of which the most impor- 

 tant is chlorophylliau absorption. 



The formation of foliary nectaries may, in principle, be due to the 

 intervention of phytophagous as well as of nectarivorous insects. The 

 tendency which ants have to tear the leaves of young peach-tree buds 

 is well known. It may be supposed that the bites of these insects 

 upon the inferior portions of the leaf caused a progressive atrophy of 

 that organ. These portions would be progressively adapted to a new 

 function — that of nectaries. The plant, thus forced to adapt itself to 

 the needs of the ants, would in this manner establish a modus vivendi 

 between itself and those insects. In place of giving up portions of its 

 foliary parenchyma, it would give them a sugary liquid. The ants 

 would find every advantage in this substitution, the liquid being more 

 easily assimilable and its collection by sucking being much more eco- 

 nomical in time and labor than the mastication of the foliary paren- 

 chyma. In return the ants would protect the plant against the attacks 

 of phytophagous creatures. 



Along the edges of the leaves of the Rosa Banksice are found perifo- 

 liary nectaries that attract great numbers of a large black ant (Campo- 

 notus pnbescens). The presence of these ants preserves the rose from the 

 attacks of a hymenopterous insect (Hylotoma rosce). We owe an inter- 

 esting experiment upon this subject to Beccari. On a branch of Rosa 

 Banksice attacked by ants he placed a branch of another rose bush 



