Jtme 20, 1878] 



NATURE 



193 



The Tailed Amphibians, hicluding the Ccecilia?is. A 

 Thesis presented to the Faculty of Michigan University 

 by W. H. Smith. (Detroit, Michigan, 1877.) 



The title of this little volume tells its ovm story. It is a 

 detailed catalogue of all the species of tailed amphibia 

 known. In addition to using the works of all the best 

 writers on this group, Mr. Smith has availed himself of 

 the specimens in his University Museum, and from these 

 has drawn up many of the descriptions and characters. 

 A number of artificial keys are given to the genera and 

 species ; the synonymic lists appear to have been worked 

 out with care, and to have been brought down to date. A 

 list of authors on the subject of the work is appended, 

 and here and there, after the diagnoses of the species, will 

 be found details of their habits, geographical distribution, 

 and development. 



LETTERS TO THE EDITOR 



[TTie Editor does not hold himself responsible for of inions expressed 

 by his correspondents. Neither can he undertake to return, 

 or to correspond with the writers of rejected manuscripts. 

 No notice is taken of anonymous communications. 



[The Editor urgently requests correspondents to keep their letters at 

 short as possible. The pressure on his space is so great that it 

 is impossible otherwise to ensure the appearance even of com' 

 munications containing interesting and novel facts.l 



Indian Rainfall 



Agreeing in the main with the views put forward by Mr 

 Archibald in his letter in Nature (vol. xvii. p. 505), I beg 

 leave to refer briefly to one or two points in which I differ from 

 him, and I hope that you will be able to find space for this note, 

 because Mr. Archibald has done me the honour of mentioning 

 my name so frequently in his letter, that I might reasonably be 

 supposed to entertain opinions identical with his own on all 

 points regarding the question under discussion. 



In the first place I would point out that the atmospheric cur- 

 rent which brings the winter rains of Northern India, whilst it 

 has nothing to do with the summer monsoon, does not descend 

 in the Punjab, as Mr. Archibald says, and then proceed east- 

 wards to the North- West Provinces and Behar, and sometimes 

 even as far as Calcutta, but blows in just the opposite direction, 

 appearing as a south-east wind over the Gangetic plain and the 

 Eastern Punjab. The place of its descent in the winter months 

 is farther south, in latitude 22° or 23° N., and thence it flows 

 northwards in almost the same manner as the summer monsoon. 



In the next place, I think the hypothesis of the approxi- 

 mately inverse variation of the winter rain, as compared with 

 sun-spots, does not necessarily postulate a corresponding inverse 

 variation of solar radiation. Such a relation I consider to be 

 highly probable, but the somewhat meagre data I was able in a 

 former communication (vol. xvi. p. 505^) to adduce in favour 

 of it were only intended to prove that the question of "solar 

 activity " was yet an open one, and that it did not follow that 

 solar radiation was most intense at times of maximum sun-spots, 

 because many meteorologists, reasoning from magnetic and other 

 analogies, assumed it to be so. The direct solution of the ques- 

 tion must be accomplished by actinometric observations, as Mr. 

 Blanford proposes, and, while it remains unsettled, it will pro- 

 bably be best to try and correlate the variations of rainfall with 

 those of some other meteorological element upon which rainfall 

 depends. I have recently been occupied with an analysis of the 

 rainfall observations of twenty stations in Northern India, embrac- 

 ing between them ii* of latitude and 24° of longitude, and extend- 

 ing over periods of from fifteen to forty-nine years, and I find 

 a remarkable coincidence between the variations of the winter 

 rainfall and those of the temperature of the tropics as given by 

 Koppen in his exhaustive paper in the Zeitschrift der oestcr- 

 reichischen Gesellschaft fiir Meteorologie, vol. viii., Nos. 16 and 

 17. When the rainfall deviations of the different stations are 

 thrown into the form of percentage variations from the local 

 mean and are then combined and the results " bloxamed," we 

 get a series of numbers which gives a curve from 1834 to 1877 

 resembling Koppen's curve very closely, when the latter is 

 extended up to 1877. The two curves not only resemble each 

 other in all their more important fluctnations, but their epochs 



of maximum and minimum approximately coincide. These 

 are : — 



Mpv \ Tropical Temperature 18427, 18547, 1865-1, 1876-3 (?) 



• 1 Winter Rain 18427, 1855-0, 1865-5, i876-9(?) 



,,. f Tropical Temperatiure 1836-9, 1847-7, 1858-4, 1874-8 

 ''""• t Winter Rain 1837-8,1848-1,1860-6,1874-7 



It would therefore appear to be highly probable that the periodic 

 variation of the winter rainfall of Northern India is caused by a 

 corresponding variation in the temperature of the tropics, which 

 determines, within certain limits, the quantity of vapoiur added 

 to the air and the direction and velocity of the atmospheric 

 currents. It appears, also, from the table, that the maxi- 

 mum of winter rainfall is attained nearly a year before the 

 minimum epoch of sun-spots, as given by Wolf, I have found 

 that this is also the case with the winter rainfall of London, and 

 Mr. Draper has shown (Nature, vol. xvii. p. 16) that the sam.e 

 relation holds good at New York. 



The co-existence of severe droughts in Hindustan with devas- 

 tating floods in Burmah and Assam, is a very strong argument 

 against the theory of Dr. Meldram that the rainfall of the whole 

 globe varies directly with the sun-spots ; but it would naturally 

 follow from the view advocated by Mr. Archibald, becau!:e, in 

 very hot years, which are approximately those of minimum sun- 

 spot, the general tendency to a cyclonic circulation of the atmo- 

 sphere round the Asiatic continent in the summer months would 

 be so intensified as partially to obliterate the smaller cyclonic 

 indraught towards Central India, which brings up a moist 

 current from the Bay of Bengal to the Himalaya and the 

 plains of Northern India. S. A. HiLL 



Allahabad, May 18 



A Twenty Years' Error in the Geography of Australia 



In almost every detailed map of Australia, including some of 

 the latest, we find, at the head of the Alligator River, in about 

 S. lat. 135°, and E. long. 133°, some such note as this : — 

 " Steep walls, 3,800 ft." This is copied from the map illus- 

 tradng "Leichardt's Journal," published in London in 1847. 

 This map was (as stated in the preface) drawn by S. A. Perry,- 

 Esq., Deputy Siurveyor-General of New South Wales, from 

 materials furnished by Leichardt, and was engraved in London 

 by Arrowsmith. As Leichardt only returned from his first 

 expedition at the end of 1845 or beginning of 1846 he could 

 have had no opportunity of correcting or revising this map. 

 Mr. James Wilson, the geologist to the North Australian Ex- 

 pedition under Mr. A. C. Gregory, having passed over much of 

 the same country, and finding the plateau nowhere more than 

 1,600 feet above the sea, came to the conclusion that Leichardt's 

 supposed statement was an engraver's or printer's error which 

 had escaped correction, and gave his reasons for this view in 

 the Proceedings of the Royal Geographical Society, vol. i. 

 p.- 230, and subsequently in the same society's Journal, vol. 

 xxviii. p. 137 (1858). Notwithstanding the extreme impro- 

 bability — almost amounting to absiurdity — of there being pre- 

 cipices of the enormous height of 3,800 feet, in a country 

 where there were no important mountains, and where Gregory,, 

 who had passed within eighty miles, and M'Douall Stuart, 

 who had passed within forty miles of the place, found nothing 

 but a moderately-elevated plateau, with ravines never exceed- 

 ing 600 feet deep, no notice appears to have been taken of Mr. 

 Wilson's correction, but the "3,800 ft." has been copied again 

 and again in works of reputation and authority. We find it even 

 in the new edition of the " Encyclopa;dia Britannica,"art. " Aus- 

 tralia," given as an e^.tablished fact in the following words : — 

 " On the north side of the continent, except around the Gulf of 

 Carpentaria, the edge of the sandstone table-land has a great 

 elevation ; it is cut by the Alligator River into gorges 3,800 ft. 

 deep." 



The curious 'thing is, however, that this marvellous pheno- 

 menon, which, if it existed, would be unapproached in Australia 

 and equalled nowhere but among the mountains of the great 

 continents, is not even alluded to in the published journal of the 

 traveller who is supposed to have discovered it ! On Leichardt's 

 map the "steep walls" are noted between the stations of 

 November 10 and 11, but in his "Journal" we find no reference 

 to anything remarkable till November 17, when he comes to the 

 head of a magnificent valley, into which he was obliged to 

 descend, and which caused him much delay and circuitous- 

 explorations en account of its steep rocky waUs estimated by 

 him to be " i,8co feet high." It is pretty clear, then, that the 



