BIOLOGIC RELATIONS BETWEEN PLANTS AND ANTS. 441 



the plant indirectly in its struggle against drought. The irritation 

 produced by the insects causes an hypertrophy of the tubercle and 

 consequent increase in the size of the reservoir. If the galleries were 

 primarily the work of ants, they must have been, on general principles, 

 unfavorable to the plant, which has, however, by progressive adap- 

 tation, finally utilized them. 



On reflection we are led to believe that the labyrinth within the 

 tubercle must be a feature very useful to the plant. It permits an 

 active circulation of atmospheric air within the tubercle, and the pres- 

 ence of oxygen may be necessary for the elaboration of certain nutritive 

 principles within its tissue. But the utility of this feature seems to be 

 of another kind. The suggestion we are about to offer has not been 

 proposed by any of those who have occupied themselves with the study 

 of these plants, yet we think it merits attention. The corky layer that 

 invests the entire surface of the tubercle is an obstacle to gaseous inter- 

 change between that body and the exterior. Such interchange can 

 only take place by means of the air that circulates in the galleries. 

 There, too, it can only be effected by the lenticels, since the internal 

 surface is also lined with a corky layer, except where these lenticels 

 are found. Now, sudden changes in the hygrometric state of the sur- 

 rounding air will be but slowly transmitted to the air of the galleries, 

 and it is this hygrometric state that regulates the interchange of water 

 vapor between the tubercle and its environment. The presence of the 

 labyrinth of galleries would then permit the plant to adapt itself more 

 readily to the hygrometric changes in the circumambient air, which 

 changes must be sudden, owing to the epiphytic situation of the plant. 

 In case of drought the plant finds in its tubercle a reserve of water, its 

 fleshy leaves transpire but little, and finally the air of the galleries is 

 nearer the point of saturation than is the surrounding air. Hence the 

 transpiration of water by the lenticels is less than it would be if they 

 were exposed to the dryness of the surrounding air. 



Some of the walls of the galleries of the Myrmecodia are smooth, others 

 (PI. XX) studded with little prominences that might a priori be sup- 

 posed to be glands for absorbing nutritive principles derived from the 

 decomposition either of the carcasses of ants (a rare case, since the 

 dead are usually dragged out of the nest), or of detritus occasioned by 

 the work of those insects. Treub has made a careful study of these 

 prominences and has shown that they are internal lenticels, differing 

 but little from the ordinary external lenticels. It is well known that 

 the function of the cellular masses forming the lenticels is to supply 

 atmospheric air to the tissue of the plant. The lenticels of Myrmecodia 

 differ from those of other plants by absence of the central aeriferous 

 passages; but all about them the files of peripheral cells are filled with 

 air, and this may compensate for the lack of passages between the 

 central files. It is also possible that certain protoplasmic-bodied 

 cells that surround the lenticels like a collar serve to elaborate and 



