544 



NATURE 



{Oct. 19, 1876 



specimens of Potentilla rej>tans, less than half an inch in dia- 

 meter, and even in unexpanded buds, were the pollen-tubes 

 penetrating the stigmas. 



I call attention to pollen-tubes, because, unless they be 

 observed, one cannot feel absolutely certain that the flowers are 

 really self-fertilised ; and even then, that fact must be associated 

 with the relative positions of anthers and stigmas, and the 

 resulting abundance of fruit. 



Another point I would mention of importance is the necessity 

 of observing the order of emergence of the whorls. The subsequent 

 rates of growth may prove a source of deception, so that it is neces- 

 sary to go back to the very earliest condition when the parts are 

 little more than papillae, and if possible even before one or more 

 of the whorls have put in an appearance at all. Now I find that 

 in conspicuous flowers, with certain exceptions, the corolla is 

 very often the last to emerge, though ultimately it attains by far 

 the largest size when adult ; that the stamens usually come directly 

 after the calyx, which, if present, is always first, acting as a 

 protecting and nourishing organ ; and that the pistil comes next. 

 Such an order results usually in protandry ; but while con- 

 spicuous species, as Stellaria Holostea, and Cardamine pratensis, 

 have the order, calyx, stamens, pistil, corolla, inconspicuous self- 

 fertilising species are often as follows : — e.si., Cerastium glome- 

 latum — calyx, pistil, stamens, corolla, and Nasturtitun officinale, 

 calyx, stamens and pistil (together), corolla. These examples, 

 out of many collected, appear to point to an important connection 

 between the order of emergence and development on the one 

 hand, and cross and self-fertilisation on the other. The connec- 

 tion between these two orders of facts I take to be, as already 

 stated, due to the fact that the energy of conspicuous flowers is 

 diverted into the corolla, which thereby delays the development 

 of the pistil ; but when the corolla is arrested, the pistil recovers 

 itself, and its growth is equal to or precedes that of the stamens, 

 the result issuing in a synchronous maturity, and consequently 

 self-pollination, George Henslow 



7, Bentinck Terrace, Regent's Park 



Wallace's " Geographical Distribution of Animals " 



Allow me to point out in Nature a few errors which occur 

 in Mr. Wallace's "Distribution of Animals," regarding the 

 extinct mammalian fauna of India. 



In the first place, there is a mistake regarding the locality of 

 the Perim Island (vol. i., p. 362, vol. ii., pp. 157 and 221), from 

 which Tertiary fossils have been obtained ; in Mr. Wallace's 

 book the Perim Island, at the entrance to the Red Sea, is the 

 one referred to, whereas the true spot is Perim Island, in the 

 Gulf of Cambay. 



There is, therefore at present no known spot to the eastward 

 of India which shows the former extension of its Tertiary mam- 

 malia into Africa and Europe, although such extension doubtless 

 existed. 



The extinct genus Enhydriodon, from the Siwaliks [e.g. vol. 

 ii., p. 200), is always referred to as Enhydrion. 



In vol. i. (p. 122) the genus Tapirus is mentioned as occur- 

 ring in the Miocene of the Punjab ; this determination is on the 

 authority of Dr. Falconer, who hastily examined a single tooth 

 (now in the Indian Museum) ; this tooth, and others subse- 

 quently found, turns out to belong to the European Miocene 

 genus Listriodon ; the only other mentioned occurrence of a 

 ibssil Tapir in India, is by Mr. Clift, who figured the symphysis 

 of a mandible [Geol, Trans., sec. ser., vol. ii.) from Burma; 

 this may, however, also belong to Listriodon. 



In vol. ii. (p. 202) the genus Ursus is mentioned as having 

 been described from the Siwaliks and the Nerbudda Valley ; it 

 has only been described from the latter locality, Hysenactos 

 being the Siwalik genus. A new species of tame Ursus has, 

 however, been obtained this year from the Siwaliks, and will be 

 subsequently described. 



In vol. ii. (p. 212) Hipparion should also be mentioned as 

 having been found in India as well as in Europe. 



At p. 228 of the same volume, it is stated that Elephas has 

 "perhaps one species Pliocene in Central India ; " in reality 

 there are two species undoubtedly from the Newer Pliocene of 

 the Nerbudda Valley, viz., E. nomadicus and E. (Stegodon) 

 insignis. 



Vol. ii., p. 240, the genus Hystrix has been fossil in the 

 Siwaliks of India as well as in Europe and America. 



I may add that, as announced in the August number of the 

 " Records of the Geological Survey of India," for the present 

 year, I have determined the existence of a ipecies of Manis (th« 



first fossil species of the genus) and of a Cetacean, vf ith other 

 new forms, from the Siwaliks. Richard Lydekker, 



Calcutta, August 27 Geological Survey of India 



The Resistance of the Electric Arc 



For the purpose of determining theoretically the best arrange- 

 ment of cells for the production of the electric light, it was neces- 

 sary to know the resistance of the electric arc. Not being 

 acquainted with any source from which this information could 

 be derived, we determined this resistance experimentally in two 

 distinct ways. 



1. The current from sixty new Grove's cells joined in series 

 (and of which the immersed part of each platinum plate was 

 about 13 square inches in area and of each zinc plate about 25 

 square inches) was used to produce an electric light with a 

 Duboscq's lamp, when a small known resistance consisting of 

 many metres of thick bare copper wire hanginjr in the air was 

 also introduced in circuit. This wire was sufficiently thick for 

 its resistance not to be sensibly altered by tlie passage of the 

 current. The difference of potentials between the carbons was 

 measured with a Thomson's quadrant electrometer, using the 

 induction plate and compared with the difference of potentials 

 between the two ends of the wire of known resistance. These 

 two measurements were made rapidly one after the other and 

 repeated very many times. Then since at any moment the same 

 current is flowing through the electric arc and the wire, the two 

 differences of potentials measured rapidly one after the other are 

 proportional to the resistances. 



The above method showed that the resistance of the electric 

 arc varied considerably even when the light appeared quite 

 steady, that the resistance was never more than 20 ohms, and 

 had an average value of about 12 ohms. 



2. On another occasion the current from eighty similar Grove's 

 cells joined in series, which had been joined up for three hours, 

 a»d used at intervals during this time for the production of the 

 light, was sent through the coils of a differential galvanometer. 

 In one circuit was a very high resistance and in the other the 

 electric arc ; each coil of the differential galvanometer was shunted 

 with a wire of small resistance. Nearly the whole current, 

 therefore, w«nt through the arc. The shunts being properly 

 adjusted to obtain balance, the resistance of the arc, as in the 

 previous case, was found to vary much but never to exceed 29 

 ohms and to equal about 20 ohms when the light was best. 



That the resistance would be larger than in the previous case 

 was to be expected since the battery contained more cells, and a 

 brighter light would, therefore, be obtained with the carbon 

 points further apart. 



At a convenient opportunity we hope to take time readings of 

 the resistance together with photographs of the light on a revolv- 

 ing band of sensitive paper in order to determine the exact resist- 

 ance corresponding with the brightest light for any particular 

 number of cells. 



The results, however, given above show that with cells such 

 as we used, and which are the common Grove's cells employed 

 in England, no attempt should be made to join any of the cells 

 in parallel circuit until at least 200 have been joined in series, 

 for since the resistance of each cell is about 0'2 ohms, 200 of them 

 would have a resistance of 40 ohms, a resistance certainly less than 

 double the electric arc for that battery corresponding with brightest 

 light, and we have shown {^Telegraphic fournal, March 15, 1873) 

 that the cells of a battery should be joined in series until the 

 battery of resistance is double the external resistance, at which 

 point the battery should be joined in two rows each containing 

 half the whole number of cells in series, and the two rows con- 

 nected in parallel circuit. W. E. Ayrton 



The Imperial College of Engineering, John Perry " 



Tokio, Japan 



Habits of Animals Transmitted to Offspring 



Breeding many horses yearly on my station, I notice, as 

 a matter of course, some of their peculiar habits. In a semi- 

 wild state on a run horses graze together in large or small com- 

 panies, which "station hands" call "mobs;" these mobs 

 wander at will over a large area of country, finding abundance of 

 good natural pasture and water. Some years since a mare 

 became solitary in her habits, always seeking one particular 

 creek ; whenever released from work she made off to her favourite 

 feeding ground by herself ; if "rounded up with a mob" she 

 ■vrould take the earliest chance that presented itself of reaching 



