450 BIOLOGIC RELATIONS BETWEEN PLANTS AND ANTS. 



accumulating, in these recesses, organic detritus capable of furnish- 

 ing assimilable soluble substances to the rootlets sheltered under the 

 concave leaves. The irritation caused on the under surface of the 

 leaves by the presence of the ants may have resulted in the augmen- 

 tation of the cavity of the organ and thus led it to take on the form 

 of a pitcher. The parasitic origin of the lodging organs is doubtful in 

 the GonchopJiyllums and the DiscMdias. 



This is not the case with the lodging organs of the Melastomacece. 

 In this family a great number of American types have leaves provided, 

 at the base of the limbus, with organs suitable for the lodgment of 

 ants. Such are the genera Lococa, Myrmedone, Majeta, Microphysca, 

 and Galophysca. It was Aublet, the old explorer of French Guiana, 

 who first remarked one of the Melastomacese that presented these curi- 

 ous features. He says, speaking of the Tococa guianensis "the leaves 

 are each attached to their stems by a little pedicle, that is at first 

 grooved on its upper surface, convex below, and set with hairs, but its 

 two sides afterwards enlarge and form a double bladder having the 

 form of a heart. This bladder is provided with two holes placed at the 

 lower part of the leaf underneath, between the two intermediary nerv- 

 ures. It is by these two holes that the ants enter and leave each divi- 

 sion of the bladder and as the stems are hollow they can penetrate 

 them also by means of openings that they make, and this is the reason 

 why some of the natives have given this plant the name of the ants- 

 nest, it being, as one may say, continually covered with them." 



He notes a similar arrangement in the Majeta guianensis, of which 

 he says "the leaves have on their under surface five longitudinal 

 nervures and numerous transversals. They are attached by a short 

 pedicle that, together with the lower part of the leaf, is enlarged in 

 the shape of a bladder, divided into two cavities by a median partition. 

 The body of the bladder is much more raised above than below. The 

 small leaves do not usually have it." This last point is of the highest 

 importance. It seems to show that the presence of ants on the interior 

 of the sac causes an hypertrophy of the inhabited leaf that is far from 

 being prejudicial to it. Of the two opposite leaves of each pair one is 

 vesicular at the base, the other and smaller one is not so. But before 

 admitting that the ants cause an hypertrophy of the foliary tissue we 

 should ascertain whether the inequality of size in the two leaves is not 

 anterior to the occupation of the vessel by these insects. It is true, 

 however, that even if this latter fact were verified we might admit 

 that primitively the ancestors of the Melastomacea? were provided with 

 equal opposite leaves having a tendency .to form a vesicle at the base 

 of the limbus. When this vesicle was once visited by the ants the 

 irritation produced by them would cause its hypertrophy and the more 

 considerable afflux of nutritive material thus occasioned would lead to 

 an increase in growth of the entire leaf. This hypertrophy would then 

 be hereditarily transmitted and would to-day be observed, even at an 



