258 



NA rURE 



\_yan. 13, 1 88 1 



On one very important point, however, M. Wesmael was in 

 error ; he states that the abdomen of the^e abnormal individuals 

 " ne contient aucun organe ; ou plutot, il n'est lui-meme qu'un 

 vaste sac stomacal." Blake even asserts that "the intestine of 

 the insect is not continued btyond the thorax," which must 

 surely be a mi- print; and also that [here is no connection 

 "between the intestine and the cloaca"! These statements, 

 however, are entirely erroneous ; and, as M. Forel has show n, 

 tlie abdomen does really contain the ufual organs, which, how- 

 ever, are very easily overlooked by the side of the gigantic 

 stomach. 



I have now the honour of exhibiting to the Society a second 

 species of ant, which has been sent me by Mr. Waller, in which 

 a similar habit has been evolved and a similar modification has 

 been produced. The two species, however, are very distinct, 

 and the former is a native of Mexico, while the present comes 

 from Adelaide in Australia. The two specie-:, therefore, cannot 

 be descended one from the other ; and it seems inevitable that 

 the modification has originated independently in the two species. 



It is interesting that, although these specimens apparently 

 never leave the nes^, and have little use therefore for legs, man- 

 dibles, &c., the modifications which they have undergone seem 

 almost confined to the alidominal portion of the digestive organs. 

 The head and thorax, antenna;, jaws, legs, &c., differ but little 

 from those of ordinary ants. 



Camponotus Inflatus, n. sp. 



Operaria. Long. 15 mill. Nigra, tarsis paliidioribus ; sub- 

 tiliter coriacea, setis cinereo-testaceis sparsis ; antennis tibii-que 

 hand pilosis ; tarsis infra hirsutis ; mandibulis punctatis, hirsutis, 

 sexdentatis ; clypeo non carinato, antice integro ; petioli squama 

 modice incrassata, antice convexa, postice plana emarginata. 



Hab. Australian. 



The colour is black, the feet being somewhat jaler. The 

 body is sparsely covered with stiff cinereo- testaceous hairs, 

 especially on the lower and anterior part of the head, the 

 mandibles, and the posterior edge of the thorax. The head and 

 thorax are finely coriaceous. 



The antenna: are of moderate length, twelve-jointed ; the 

 scape abjut one-third as long as the terminal portion and some- 

 what bent. At the apex of the scape are a few short spines, 

 bifurcated at the point. At the apex of each of the succeeding 

 segments are a few much less conspicuous spines, which decrease 

 in size from the ba-al segments outwards. The antenna is also 

 thickly clothed with short hairs, and especially towards the apex 

 with leaf-shaped sense-hairs. The clypeus is rounded, with a 

 slightly developed median lobe and a row of stiff hairs round the 

 anterior border ; it is not carinated. The mandibles' have six 

 teeth, those on one side being rather more developed and more 

 piointed than those on the other. They decrease pretty regularly 

 from the outside inwards. The maxillce are formed on the usual 

 type. -The maxillary palpi are six-jointed, the third segment 

 being but slightly longer than the second, fourth, or fifth ; while 

 in Myrmecocystus the third and fourth are greatly elongated. 

 The segments of the palpi have on the inner side a number of 

 curious curved blunt hairs besides the usual shorter ones. The 

 labial palpi are four-jointed. The eyes are elliptical and of 

 moderate size. The ocelli are not developed. 



The thorax is arched, broadest in front, without any marked 

 incision between the meso- and metanotnm ; the mesonotum 

 itself is, when seen from above, very broadly oval, almost 

 circular, rather broader in front and somewhat flattened behind. 

 The legs are of moderate length, the hinder ones somewhat the 

 longest. The scale or knot is heart-shaped, flat behind, slightly 

 arched in front, and with a few stiff, slightly diverging hairs at 

 the upper angles. The length is about two-thirds of an inch. 



ON THE THERMIC AND OPTIC BEHAVIOUR 



OF GASES UNDER THE INFLUENCE OF 



THE ELECTRIC DISCHARGE 1 



pROF. E. WIEDEMANN has undertaken an exact calori- 



metric investigation of the electric discharge through gasi s, 



and in spite of the serious difficulties which he had to encounter, 



he has already obtained valuable and important results. As a 



source of electricity. Topic r's machine was used ; but we must 



refer to the original paper for all details of experimentation. 



Three series of observations were made. In the first the total 

 heat generated in a given time in the whole vacuum tube w.as 

 measured. In the second series the capillary part only was 

 ' By Eilhard Wiedemann. {}Vied. Ann., x. p. 202.) 



examined, and in the third the thermal behaviour of the regions 

 in the neighbourhood of the electrodes was investigated. The 

 result of the first series is summed up as follows : — With de- 

 creasing pressure the total quantity of heat generated at first 

 decreases, reaches a minimum, and then increases again. In 

 hydrogen the amount of heat generated is smaller than in 

 atmospheric air. 



A smaller amount of heat developed corresponds to a larger 

 number of discharges in a given time, and hence to a smaller 

 potential at the moment the discharge begins to pass. The 

 results of Prof, Wiedeiiiann are therefore, as he p lints out, in 

 accordance with those of Messrs. De la Rue and Hugo Muller, 

 who found that the difference of potential necessary to cause a 

 discharge passes through a minimum as the pressure decreases. 



Somewhat more complicated results were obtained when an 

 air-break was introduced into the circuit. In that case the air- 

 break determines the difference of potential necess.ary to produce 

 a discharge ; but if the whole quantity of electricity would pass 

 suddenly when that potential has been reached, and before it has 

 had time to sink, the amount of heat generated would be inde- 

 pendent of the pressure in the vacuum tube. This however is 

 not the case ; but the result is intermediate between that ob- 

 tained when no air-break exists, and that which would be 

 obtained on the above supposition. 



The following re-ults were obtained in the experiments in 

 which the capillary part of a vacuum tube only was introduced 

 into the calorimeter ; — 



1. The heating effect in capillary tubes at pressures above 

 I mm. is almost independent of the quantity of electricity pass- 

 ing with each discharge, and nearly proportional to the total 

 amount of electricity which passes. 



2. The heating effect is almost the same whether the positive 

 or negative electrode of the tube is connected w ith the machine 

 (the other electrode being connected with the earth), although 

 the number of discharges passing in a given time is dillerent. 



3. With decreasing pre-sure the heat generated decreases very 

 rapidly without passing through a minimum. 



4. The heating effect is independent of the shape of the elec- 

 trodes. Some results obtained by Prof. G. Wiedemann, who 

 had found that in tubes of different widths the same amount of 

 heat is generated by the same current, were confirmed. 



Calorimetric measurements made near the electrodes showed : 



1. The heating effect near the positive electrode decreases 

 with decreasing pressure rapidly. At very low pressures a small 

 increase is sometimes observed. 



2. The heating effect near the negative electrode decreases first 

 with decreasing pressure, and then increases rapidly. 



The heating effect near the positive electrode shows some 

 anomalies when an air-break is introduced, the amount of heat 

 generated being considerably increased. 



Some measurements were reduced to an absolute scale, and 

 showed that the total amount of heat generated is very large. 

 Taking account of the number of discharges, and assuming that 

 after each discharge the gas returns to its original state, the 

 temperature in the capillary part of the tube must have been 

 about 2,000° C. at 15 mm. pressure, and about 1,100° C. at 

 5 mm. pressure. If the width of the tube was increased ten 

 times, the temperature would only be about ICO° C, and this 

 confirms the result obtained by Prof. Wiedemann in a former 

 investigation, that gases may become luminous under the 

 influence of the electric discharge at a comparatively low 

 temperature. 



In another part of the paper Prof. Wiedemann treats of a very 

 important problem. When his tubes were filled with hydrogen, 

 and an air-break was introduced in the circuit, the spectrum of 

 the luminous gas changed suddenly at a given point. According 

 to a now generally accepted hypothe^is this change of spectrum 

 is always accompanied by a change in the molecular constitution 

 of the gas ; and it is to be expected therefore that heat is either 

 absorbed or given out by a gas w hen its s pectrum changes. This 

 heat Prof. W'iedemann has endeavoured to measure. Let us 

 imagine, for instance, that the current has to do the work of de- 

 composing the molecules of a gas. The moment the discharge 

 has passed, recomposition will take place, and the heat then 

 generated was measured by Prof. Wiedemann. Some of the 

 suppositions on which the calculations are based might require 

 further investigation, but the assumptions made are supported, 

 and to a certain extent proved by the fact that the heat necessary 

 to change the band-spectrum into the line-spectrum was found to 

 be independent of the pressure and cross-section of the tube. It is 



