392 



NA TURE 



[August 24, 1893 



its fangs into ilself, when confined in a gallon jar containing 

 water, which was inverted at intervals in order to drown it, 

 is open to question as to its conclusion that it was a case of 

 " deliberate suicide," for the following reasons :— 



(i) That it was after " the snake ceased any attempt to rise 

 to the surface of the water in the jar," that the blow was 

 struck. The snake then being wholly beneath the water, 

 would die from drowning, and not from the self-inflicted 

 wounds caused by its poisoned fangs. 



(2) That it has been proved by experiment by Dr. Weir 

 Mitchell that the venom of the rattlesnake is of no effect upon 

 itself, when introduced into any wound in its body. I speak 

 from memory of an article which appeared in the Atlantic 

 Monthly some few years ago. That self-insertion of the poison 

 would make any difference is not likely. 



Drowning (by the act of others) and not self poisoning (or 

 suicide) I take to be the cause of death in the case described. 



Halifax, August 15. W. H. Wood. 



Numerous Insects Washed up by the Sea. 



The phenomenon referred to under the above heading in 

 your issue of August 1 7 may be in part accounted for by the 

 fact that on August 7, at many spots in the neighbourhood of 

 Godalming (S.W. Surrey), the air was thick for several hours 

 with swarms of winged ants. The direction of the wind was 

 from the north-west, force moderate. Assuming the like to have 

 taken place at other places, it is quite possible that large num- 

 bers of ants may have been carried out to sea and drowned from 

 this region of Surrey and Hampshire. 



Hunstanton, August 19. Oswald H. Latter. 



THE FUNGUS GARDENS OF CERTAIN SOUTH 

 AMERICAN ANTS. 



ONE of the most interesting papers that has appeared 

 during the present year, whether considered from 

 the point of view of general biology or of mycology, is 

 that which has recently been published by Herr Alfred 

 Moller, nephew of Dr. Fritz Miiller.' The work was 

 carried out at Blumenau during the years 1890-92, and 

 presents a clear and thorough investigation into the 

 habits of the leaf-cutting ants and their remarkable 

 custom of cultivating and feeding upon certain fungi. 

 The work is introduced by a quotation from Thos. Belt's 

 " Naturalist in Nicaragua," where the author, speaking 

 of the leaf-cutting ants, states : " I believe .... that 

 they are in reality mushroom growers and eaters." This 

 statement Moller fully proves in the work before us. The 

 first portion, forming the bulk of the work, is given up 

 to the consideration of the fungu; gardens of the leaf- 

 cutting ants, and is divided into ten sections. 



I. The species of the leaf-cutting ants and their activity 

 outside of the ^t-j/— Belt's description- of the Nicar- 

 aguan ants is quoted, and the differences between them 

 and those of Blumenau are pointed out. The chief point 

 of difference is that the latter form very narrow streets, 

 travelling only in single file, and that their nests occur 

 both in the forest and in the open. The commonest 

 species is Atta {Acroinynnex) discigera, Mayr, whose 

 workers are never more than 65 mm. long. Almost as 

 common is A. hystrix, Latr., whose workers reach a 

 length of 9 mm. Rarer than these are A. coronata, 

 Fabr., and a doubtful form, which Moller terms Atta IV. 



A minute description is given of a street of A. disci- 

 gera, which was 26 metres long and about 15 cm. wide 

 and high, roofed in in parts wherever possible. It led 

 to a number of small Cupheas, whose leaves the ants 

 were cutting. In the street could be seen a procession 

 of loaded ants going towards the nest, and others empty- 

 handed, going in the opposite direction. Some of the 

 large workers run up and down the road unloaded, and 



l"Die Pilzgjirten einigir siidamcrikanischer Amei?en." Heft 6 of 

 Schimper*s " B jtanische Mitthcilungenaus denTropen." (Jena: G. Fischer, 



»8?3.) 



Naturalist in Nicaragua," p. 71. 



NO. 1243. VOL. 48] 



act as road-menders if any accident happens to a part of 

 the track. Other very small workers, which do not cut 

 leaves, may also be seen carried upon the backs or even 

 upon the loads of the actual leaf-cutters. An ant carry- 

 ing a peculiarly shaped piece of leaf was watched from 

 end to end of the track, and travelled the 26 m. in 70 

 minutes. The load was twice as heavy as itself. 



The other species of the Atta have very similar streets. 

 A. hystrix appears to work only at night. 



The jaws of the ants are very strong, with serrated 

 edges, and clash together laterally. The ant begins at 

 the edge of a leaf, and cuts out a piece in about five 

 minutes, revolving on one of its hind legs as a centre. 

 W^hen the piece is almost freed, the ant goes on to the 

 main portion of the leaf, cuts through the last piece 

 uniting it with the severed portion, drags up the latter, 

 balances it on edge between its forelegs, and then, grasp- 

 ing it with its jaws, lifts it up above its head, so that the 

 centre of gravity of the load is above the ant itself. It 

 then marches off, down the stem, to the base, over the 

 ground to the end of the street, and along this to the 

 nest, travelling at a very uniform speed, and never letting 

 go its load. The weight thus carried was found, on an 

 average, to be twice that of the ant ; but many were 

 found carrying heavier loads, even as much as ten times 

 their own weight ! A street of A. coronata was watched 

 for fifteen minutes, during which time 217 ants passed, 

 carrying 3 grammes of leaves. 



2. The Nests of the Ants, and the Fungus Gardens.— 

 The nests of ^. hystrix and A. discigera are usually below 

 the surface of the soil, but covered, wherever necessary, 

 with a thick mass of withered pieces of leaves and twigs, 

 &c. They may be as much as \\ metres in diameter. 

 In the nests of all four species there is found, filling up 

 the interior, a curious grey spongy mass, full of chambers, 

 like a coarse sponge, in which the ants may be seen 

 running about, and in which, here and there, occur eggs, 

 larvae, and pups. This is the fungus garden, termed by 

 Belt "ant- food." It is separated from the roof and 

 lateral walls of the nest by a clear space. The walls 

 and roof are much thicker in winter than in summer ; 

 one nest examined had a roof 25 cm. thick and wall 

 40 cm. Photographs are given in the original paper, 

 showing the appearance of the mushroom garden. 



3. Investigation of the Gardens. The Kohlrabi 

 clumps.— T\\t garden consists of two parts, differently 

 coloured, but not very sharply marked off from one 

 another. The older part is yellowish-red in colour ; the 

 newly-built portions, forming the surface of the garden, 

 are of a blue-black colour. It is this part which is of 

 the greater importance to the ants. 



The garden is found, on examination, to consist of an 

 immense conglomeration of small round particles of not 

 more than -5 mm. in diameter, of a dark green colour 

 when quite fresh, then blue-black, and finally yellowish- 

 red. They are penetrated by, and enveloped in, white 

 fungus hyphae, which hold the particles together. These 

 1 hypha: are similar throughout the nest. 

 ! Strewn thickly upon the surface of the garden are seen 

 ' round white bodies about -25 mm. in diameter ; they 

 always occur in the nests, except in the very young portion 

 ' of the gardens. They consist of aggregations of pecul.ar 

 swollen hyphae,and are termed by Moller the " Kohl-rabi 

 clumps." The hyphs swell out at the ends into large 

 spherical thickenings, about 10-24 ix in diameter (the 

 ordinary hyphas are 5-8^ thick), filled with richly vacuo- 

 lated protoplasm like the ordinary hypha;. These clumps 

 of" Kohl-rabi" areonly found on the surface of the garden, 

 and form the principal food of the ants. A microscopic 

 examination of the particles of which the garden is com- 

 posed shows that they contain remains of leaves ; bits of 

 epidermis, stomata, spiral vessels, &c., occur in them. 



4. The Importance of the Garden to the Ants.— Ua. nest 

 be broken into and the garden scattered the ants collect 



