DONISTHORPEA. 219 



incline also on the opposite side. I have verified, upon thousands 

 of these ant-hills, this observation of the shepherds. I found a 

 trifling number of exceptions ; but only in those instances, where 

 these hillocks had been deranged by men, or other animals. They 

 do not preserve this form in the plains, where they are more ex- 

 posed to such accidents." 4 



These nests have been called ants' nest compasses, and Tissot 33 , 

 and more recently Linder 37 , have confirmed these observations. 

 The latter has shown that the elongate shape of the mounds is due 

 to the fact that the ants keep extending them in an easterly direc- 

 tion in such a manner that only the extreme easterly and highest 

 portions are inhabited by the insects. Wheeler has observed a 

 similar and equally striking orientation of the mounds of Formica 

 argentata in the sub -alpine meadows of Colorado [Journ. Psychol. 

 Neurol. 13 424 (1908)] ; and in 1910 I carried a pocket compass with 

 me wherever I was investigating ants' nests, and whenever it was 

 possible to observe their orientation I found they invariably faced 

 the east [Ent. Rec. 23 11 (1911)]. 



Sometimes the larger workers of flava are darker in colour, and 

 this is especially the case in old colonies, where many of the ants 

 will be found to be large and dark. In 1902 at Rossbeigh, Co. 

 Kerry, I found several colonies of flava, all the ants of which were 

 more or less brown in colour, but they were unfortunately described 

 by Saunders as a race of nigra as follows : " Lasius niger 

 Linn. Pale race with tibiae without exserted hairs, but much 

 larger than ordinary alienus." 30 In a colony of D. flava at Bletch- 

 ington on May 14th, 1913, in which no queen could be found I 

 captured several very large dark workers, many times larger than 

 ordinary workers 45 which measure 4'8 mm. in length and may be 

 called macrergates. 



This is a rather weak ant, and is unable to see very well ; it uses 

 its antennae to find its way, and its workers hardly ever appear 

 above ground, though a few may sometimes be seen, after rain, 

 adding to the height of their mound. When attacked by other 

 ants they block up the entrances to their nest, retiring to the 

 chambers underground, and are generally defeated. 



Landois found a structure in the worker of this species similar 

 to that he observed in D. fuliginosa which he considered to be a 

 stridulating organ, and D. Sharp describes it as a very short space 

 or ring extending all along the base of the dorsum in the third 

 abdominal segment covered with rather coarse irregular lines. 

 The line of demarcation between this and the general surface of 

 the dorsal plate is abrupt, and there is an extreme difference of 

 texture between the space uncovered and that covered by the 

 overlapping part of the preceding segment 24 . 



Very many root Aphidae live in the nests of D. flava, indeed 

 Andre says this ant lives exclusively on the liquor furnished by these 



