August 4, 19 10] 



NATURE 



159 



T 



INDIAN PALEONTOLOGY. 



HE Geological Survey of India continues to publish 

 well-illustrated and exhaustive memoirs on the fossil 

 invertebrate faunas of the region with which it deals. Two 

 more on the Himalayan Trias have lately appeared, and 

 are of much interest for study in connection with recent 

 work on the Triassic fossils of other areas. The first 

 memoir (Palaeontologia Indica, ser. 15, vol. vi., No. i, 

 1909), on the Lower Triassic Cephalopoda from Spiti, 

 Malla Johar, and Byans, was begun several years ago by 

 the late .\. von Kraftt, who collected much of the material. 

 It has now been revised, completed, and brought up to date 

 by Prof. C. Diener. It begins with a synopsis of the 

 marine Lower Triassic formations of the Himalayas, which 

 are proved to constitute a remarkably complete series. 

 The detailed descriptions of the fossils which follow show 

 fhat at least four distinct and successive faunas occur in 

 the rocks of the district under consideration. Of these, 

 the lowest or earliest is perhaps the most interesting, 

 because it seems to represent the dawn of Triassic life in 

 the sea. It is noteworthy for the complete absence of the 

 numerous types of Palseozoic Brachiopoda, which are the 

 predominating element in the Permian rocks of the Salt 

 Range and the Himalayas. Both in the Alps and in the 

 Himalayas the Permian and Trias are connected by an 

 uninterrupted sequence of sedimentary deposits. The 

 second memoir, by Prof. Diener {loc. cit., No. 2), is more 

 special, treating of the fauna, chiefly Cephalopoda, of the 

 Traumatocrinus Limestone of Painkhanda. He returns to 

 a discussion of the age of this limestone, and shows that 

 enough of its ammonites are identical with (or closely 

 allied to) species found in Europe to justify its correlation 

 with the Julie horizon, or zone of Trachyceras aonoides. 



Another memoir just received from the Geological Survey 

 of India, though dated 190S, contains a valuable descrip- 

 tion of the Devonian faunas of the northern Shan States 

 by Mr. F. R. Cowper Reed (Palaeontologia Indica, n.s., 

 vol. ii., No. 5). The fossils are chiefly corals, bryozoa, 

 and brachiopoda, with only few representatives of other 

 groups, but they constitute the richest collection of 

 Devonian age hitherto described from south-eastern .Asia. 

 Most of them were obtained from Padaukpin, and many 

 appear to be identical with European species which 

 characterise the lower part of the Middle Devonian. The 

 marine faunas of Middle and Upper Devonian times prove 

 to have been remarkably cosmopolitan, but in all casss, 

 as at Padaukpin and other places in eastern Asia, there is 

 also a local element giving them a special character. 



DUTCH METEOROLOGICAL ]VORK IX THE 



EAST^ 

 (l) \ArE have to chronicle the issue of a new set of meteor- 

 ological charts for the part of the Indian Ocean 

 around Cape Guardafui. It is issued by the Meteorological 

 Institute of the Netherlands to replace a set of similar 

 charts published in 1888 which is now out of print. The 

 observations are now sufficiently numerous to justify the 

 subdivision of the restricted area under discussion into 

 squares measuring 12' by 12', so as to bring out variations 

 over short distances. 



Special attention has been given to currents. Nearly 

 4000 observations, extending over the period 1S88-1908, 

 have been useH, and these are all based on astronomical 

 observations made at intervals of six or eight hours. The 

 results are represented by " current roses," giving for each 

 subsquare the mean velocity observed from each of sixteen 

 directions. This method of representation gives an excel- 

 lent idea of the varying nature of the currents ; thus some 

 of the roses quite near to Cape Guardafui are very nearly 

 symmetrical stars, showing that currents from all direc- 

 tions may be encountered. We rather miss an indication 

 of the number of observations used in computing each 

 vector. Some must be based on very few observations, 

 while others represent the mean of a considerable number 



1 (i) Koninklijk Nederland^ch Meter>rologisch Institimt, No. los. 

 Oceanographische en MetcorologiscKe Waamemingen bij Kaap Guardafui. 

 Pp. 38. (Amsterdam : H. G. Bom, n.d.) Price 6.00 florins. 



(2) Regenwaarnemingen in Nederlandsch-Inlie. Dertigste laareang 

 iqoS. Heel !., nagelijksche Regenv,il. Pp. vi + 392. Deel ii., Uitkomsten. 

 Pp. xn-hiQo. (Batavia : Landsbrukkerij, 1909.) 



of records, and we cannot help thinking that it would be 

 useful both to the mariner and to the student to be in a 

 position to weight the results. 



The charts which follow give for each month and for 

 each subsquare the averages for wind, pressure, temperature 

 ot the air, and temperature of the water. In the last the 

 extraordinary low temperature of the surface water off the 

 coast south of Cape Guardafui during the south-west mon- 

 soon is well shown. In July we find a reading of iS-S" C. 

 (65-8° F.) slightly south of Ras Hafun, while in the Gulf 

 of Aden, slightly west of Guardafui, a subsquare has a 

 mean temperature of 30° C. (86° F.). Nevertheless, there 

 IS a caution in the introduction against relying on low 

 surface temperatures to give warning of the proxim'ty of 

 land during hazy weather, for warm water is occasionally 

 encountered south of Guardafui. 



It has been found necessary to represent the monthly 

 results for each element on a separate chart. Our first 

 impulse when dealing with charts of this nature is to com- 

 pare the results for different elements, and for this it is 

 very laborious to have to refer to five different charts bound 

 in different parts of the volume. We admit that there is 

 danger of overcrowding, even if different colours are used 

 for different elements, but it is a great advantage to be 

 able to survey the complete data for a month with a 

 minimum of cross-references. 



(2) The rainfall volumes for the Dutch possessions in 

 the East Indies form the thirtieth issue of the series. The 

 first of the two volumes for 190S gives daily observations 

 of rainfall for 272 places. In the second volume we have 

 statistics of the number of rain-days, greatest rainfall in 

 a day, and comparisons with averages, and also a dis- 

 cussion of the records of eight autographic gauges. The 

 publication, especially the first volume, which gives the 

 names of all the observers for each month separately, seems 

 at first sight unnecessarily detailed, but in a country where 

 the rainfall is of such great economic importance such 

 details are necessary, and do much to ensure accuracy. 



PRACTICAL SPECTROSCOPY. 

 A NOVEL and very compact form of mounting for 

 ■^ concave gratings is described by Mr. Albert Eagle in 

 \o. 2, vol. xxxi., of the Astropliysical Journal. Such a 

 mounting has been erected, and found very satisfactory, in 

 the spectroscopic laboratory of the Royal College of Science, 

 and it is the experience obtained from this that has led 

 to the publication of the details for general use. 



In the Rowland form of mounting in general use there 

 are serious disadvantages, the chief of which is that a 

 large and darkened room must be devoted solely to the 

 spectrograph when in use ; the difficulty of efficient tempera- 

 cure control is also a serious one. In the new form most 

 o( the disadvantages are eliminated, and no serious new 

 ones are introduced. The whole spectrograph for a lo-foot 

 grating is, in the new form, contained in a box 11 feet 

 I inch in length, 25 inches broad, and 22 inches deep, and 

 tne plates reproduced iii the paper prove conclusively the 

 efficiency of the apparatus under the ordinary conditions 

 of laboratory work. On a photograph of the cyanogen 

 band at \ 3883, taken in the fifth oider with an exposure 

 o' forty minutes, lines only 005 .Angstrom are distinctly 

 resolved. The temperature is maintained constant by 

 lagging the double walls of the camera tube with slag 

 wool, and the reproduction of part of a first-order iror. 

 spectrum, given four separate exposures of ten seconds at 

 intervals of an hour, shows how efficient it is ; this spectrum 

 was taken without any special precautions whilst other 

 work was proceeding in the well-lighted laboratory as 

 usual, and yet the close pair of lines, separated by only 

 o-ii8 Angstrom unit, at \ 4240 is resolved. Other 

 advantages claimed tor the new mounting are its com- 

 paratively low cost, its rigidity, a slightly increased dis- 

 persion, the use of higher orders than in the Rowland 

 mounting, and the fact that the orders on either side of 

 the normal may be employed. .Against these are to be set 

 two or three apparent disadvantages, of which, at first 

 sight, a slight departure from normality appears to be the 

 most serious ; but, as pointed out by the author, an observer 

 always has to construct a curve of errors when reducing 

 observations, and such a curve would include this slight 



NO. 2127, VOL. 84] 



