1906.] Wliccler, Relations of Ants to Plants. 407 



inhabited by colonies or parts of colonies at the acme of their 

 prosperity and development, are completely or almost completely 

 free from grass. There can be no doubt that this vegetation 

 usually succumbs to the sharp mandibles of the ants. A 

 number of nests of this description, appearing as compact earthen 

 mounds, covered only with a small quantity of vegetable detritus, 

 were seen in a large clearing near Scotch Plains. Four of these are 

 shown in PI. LXIV. They were all smaller than any of those meas- 

 ured for the above table. 



4. This stage is characterized by a growth of moss which first 

 makes its appearance in a narrow band around the extreme base of 

 the mound just above a shallow, moat-like depression about 20 cm. 

 in diameter which often encircles the nest and separates it from the 

 surrounding surface. This moss is Ditrichum pallidum,^ a species 

 which is too small and delicate to interfere with the activities of the 

 .ants. In the course of time it is partly or completely supplanted by 

 a larger and more vigorous species {Polytrichum commune), which 

 begins to form a densely tufted zone around the base of the mound. 

 This moss is rarely found in the woods near Scotch Plains except on 

 the ant-hills, and I was at a loss to account for its singular distri- 

 bution till Mrs. Britton informed me that it is preeminently a bank- 

 loving species. As the soil of the woods is very level, the plant 

 naturally takes to the only bare elevations in the vicinity. It is not 

 improbable that the Ditrichum may in some way prepare the soil for 

 the growth of the Polytrichuin. PL LXV represents the largest nest 

 in the colony (No. 8 of the above table) , which had a zone of Ditrichum 

 10-20 cm. broad around its base, but too delicate to show in the 

 photograph. Plates LXVI, LXVII, and LXVIII show that the 

 Polytrichum zone continually widens until only a small area at the 

 summit of the mound, the umbilicus, is left uncovered. In Plate 

 LXVII it has invaded a nest which retained a well developed grass 

 zone, but this is evidently destined to disappear before the advancing 

 moss. Nests over which the Polytrichum carpet is far advanced are 

 invariably depauperate. The ants are few in number and seem to 

 be rather inactive, a condition undoubtedly attributable to the 

 growth of the moss. This plant has very hard, tough stems and 

 roots, which the ants are unable to sever, so that they cannot dispose 

 of it as they do of grass and other plants. It chokes up the basal 

 zone of entrances, or at any rate grows so densely as to make them 



'For kindly identifying the mosses mentioned in this paper I am indebted to the well known 

 bryologist Mrs. Elizabeth G. Britton. 



