212 LSoptember, 



tussock grass, each tussock forming a little hillock from one to two feet high, the 

 ground between and below tiie tussocks being wet and mossy. On the top of one 

 of these tussocks was a nest, in the shape of a cone, composed of very small bits of 

 dried grass. It was about 9 inches high, and 4—5 inches wide at the base, the 

 whole supported by the blades of tussock grass on the sides, while some of the 

 blades sprang out from the top, forming a sort of pillar in the middle of the nest. 



The ants looked like ordinary Formica fusca, but were slightly smaller and 

 much more shining. 



Mr. Saurders, to whom I submitted them, considers them to belong to the race 

 gagates of F. fusca. I think it is probable that the peculiar shape of the nest may 

 have been due to the nature of the surroundings, as the ants could not build except 

 on the tops of the hillocks, and in wet weather these would form so many islands 

 in a miniature lake. — G-. Arnold, Royal College of Science, South Kensington : 

 August^ 1905. 



[This form of Formica fusca is an interesting one, as it is certainly rare in 

 Britain ; it is identical with the form which I considered to belong to the race 

 gagatex in my " Hymenoptera Aculeata," and which the Rev. W. Farren White 

 described as a new species under the name glahra in " Ants and their Ways." 

 I sent two of Mr. Arnold's specimens to Prof. Forel for his opinion, aud he returns 

 them as F. fusca race gagates " une peu fuscoide." — E. Saunders.] 



Hgmenopfera and Hemiptera in the Mendips. — From June 22nd to July 13th 

 I collected at Glastonbury and Winscombe. The weather was brilliant and every- 

 thing seemed to be in a mood which one would have thought most attractive to the 

 Aculeate Hgmenoptera ; but although I was constantly searching the most favour- 

 able localities I found practically nothing worth recording. A single Agenia 

 variegata g, Crabro capitosus 1 ?, 2 Passalaecus monilicornis, a few Odynerus 

 melanocephahis, and 1 SteHs aterrima were the only species not actually common. 

 Not only were the number of species few, but even individuals of common ones 

 were scarce. On one occasion, about noon of the 5th July, I searched a bank 

 facing nearly due south, unusually gay with flowering plants such as ought to 

 attract any respectable bee, amongst them being the following : Ranunculus, 

 Selianthemum , Hypericum, Medicago ItipuUna, Trifolium pratense, Lotus, Poten- 

 tilla, Agrimonia, Heracleum, Daucus, Galium verum, Chrysanthemum leucanthemum, 

 Achillea, Senecio, Centaurea, Hypochceris, Crepis, Lapsana, and Prunella, all in 

 abundance; notwithstanding this combination of flowers, a few Bombus agrorum^ , 

 Apix, and one Halichis leucozonius ? , were the only visible Aculeates. 



No doubt the nature of the soil (limestone) is unfavoui-able to Hymenoptera, 

 but I do not remember a similar experience anywhere. 



Hemiptera were represented by one or two better species, Macrocoleus hortu- 

 lanus and Asciodema fieberi being the best, the former common on Helianthemum 

 flowers and the latter rather rare on Ulmus montanus. Heterocordylus unicolor 

 swarmed on Genista tinetoria in several localities. — Edward Saunders, St. Ann's, 

 Woking : August \st, 1905. 



