October 17, 1901 



NA TURE 



599 



fact that the zero of attitude does not correspond with the 

 axis of colhmation of the telescope. This displacement 

 of the zero line has been necessitated by the desire to 

 make the instrument available for the measurement of 

 differences of altitude amounting to 30^, and to get the 

 resulting displacement for such elevation it was necessary 

 to use more than the semidiameter of the field. The 

 author discusses the amount of error likely to arise from 

 this cause and puts the result in a tabular form. Very 

 full descriptions of the method of adjustment are given 

 and some very practical remarks are made on the method 

 of using the apparatus. 



To judge by the examples that the author has given, 

 the instrument should prove very useful in the hands of 

 an expert. These e.xamples show that in the measure- 

 ment of a distance of 250 m. an error of about 06 m. 

 may be apparent, while the average error in elevation 

 over the same distance, and in which the variation of level 

 amounts to about ± 7', will amount to a few centimetres. 



OUR BOOK SHELF. 

 Resulls of Meteorological Observations made at the Rad- 

 cliffe ObservoJory, Oxford, in the eioJit years 1892-99. 

 Edited by Arthur A. Rambaut, M. A. (Dubl. and Oxon.), 

 Sc.D., F.R.S., Kadcliffe Observer. Vol. .xlviii. Pp. 

 xxiv + 245. (Oxford : J. Parker, igoi.) 



The publication of a collection of meteorological obser- 

 vations made in 1892 may at first sight appear somewhat 

 belated, and as giving promise of but little interest. But 

 observations such as the greater part of those contained 

 m this volume serve two purposes. There is first of all 

 the immediate application of knowledge concerning the 

 atmospheric variations whose usefulness is shown in 

 weather prediction and similar purposes. Some may 

 think that this is the main, if not the only, outcome of 

 meteorological inquiry. But, apart from all ephemeral 

 interests, the maintenance of a continuous record of the 

 behaviour of the atmosphere is of great importance. The 

 study of climatic oscillations throughout long periods is 

 a study that is likely to be attended with great advantage 

 and instruction. The long, costly and laborious series of 

 observations, that are so carefully prosecuted at so many 

 stations, can only be justified by their use in investigations 

 which aim at the primary causes of atmospheric disturb- 

 ance. The records of the Radcliffe Observatory hold a 

 deservedly high place in such series, both for accuracy 

 and for length of time during which they have been un- 

 interruptedly pursued, and for the purposes of scientific 

 meteorology the value of the present volume is undi- 

 minished by the length of time that separates us from the 

 earlier observations. It will take its place among many 

 worthy companions and hand on the history of the 

 variation of climate to those who have the skill to read it. 

 A feature of great additional interest is given to the 

 present Volume by an inquiry into underground tempera- 

 tures at various depths by means of platinum-resistance 

 thermometers. This inquiry was originated under the 

 direction of the late Mr. E. J. Stone, and has been 

 vigorously prosecuted by the present director. The ther- 

 mometers are placed at depths varying from six inches to 

 ten feet ; a greater depth, which was originally contem- 

 plated, being found impracticable owing to the presence 

 of water in the soil. The present inquiry is limited, but 

 precise. It concerns itself with the variation of tempera- 

 ture in dry gravel ; and the thermal conductivity of a 

 water-logged stratum, or of one greatly differing m con- 

 stitution from that here investigated, does not come into 

 consideration. The main conclusion to which the Rad- 

 cliffe Observer is led in this investigation into the physics 

 NO. 1668, VOL. 64] 



of the earth's crust is, that the annual variation of tem- 

 perature is reduced to o'-i F. at a depth of 453 English 

 feet, and to o' 01 F. at 66 English feet. The s'emi-annual 

 wave has the same limits at 21 '4 and 36 feet, respectively. 

 The temperature curves for the separate months of the 

 year on which this result is based are shown graphically 

 in a plate possessing many features of interest. 



But of equal, if not of greater, importance is the inquiry 

 into the accuracy of the thermometers themselves and 

 their suitability for such investigations. One gathers that 

 although very considerable difticulty was experienced at 

 the outset, and not unnaturally with a novel kind of 

 apparatus, these thermometers have stood the test with 

 great satisfaction and proved themselves more trust- 

 worthy and more convenient than the long-stemmed 

 spirit thermometers ordinal ily employed m similar 

 researches, and against which soine obvious objections 

 can be urged The main difficulty in the use of the 

 platinum-resistance thermometer seems to arise from a 

 damp atmosphere affecting the connecting wires and 

 impairing the insulation, but with sufficient precaution 

 the recording apparatus is most sensitive and permanent. 



The Telephone System of the British Post Office. By 

 T. E. Herbert. Pp. xi 4- 218. (London: Whittaker 

 and Co., 1901.) Price ^s. 6d. 

 Mr. T. E. Herbert describes the book before us as a 

 practical handbook, and, from certain expressions used 

 in the second chapter, he seems to be one of those prac- 

 ticians who have not overmuch sympathy with theoretical 

 workers. It is not perhaps to be wondered at, therefore, 

 if the preliminary chapters of his book, dealing with the 

 fundamental prmciples of sound, electricity, magnetism, 

 and telephony are handled in a very unsatisfactory man- 

 ner. We are afraid that a reader, if he has not already 

 acquired a thorough knowledge of the subject, will be 

 liable to form erroneous impressions. Thus, to give one 

 example, Mr. Herbert stales that in an induction coil 

 "the E.M.F.'s generated in the secondary coil are 

 directly proportional to the current variations in the 

 primary." .Again, the description of the action of the 

 Bell transmitter is, we are inclined to think, incorrect, 

 as the same mistake is made here of not properly allowing 

 for the time taken over a \ ibration of the diaphragm. 



The greater part of the book is devoted to a detailed 

 consideration of the apparatus and connections used by 

 the Post Office. This would have been greatly improved 

 if more care had been taken with the diagrams. It is a 

 great pity that in a book of this kind, where clearness in 

 the illustrations is so important, the lettering should be in 

 some cases so small as to be unreadable. It is to be re- 

 gretted, too, that such words as "nextly '' and " inoxidible" 

 are allowed places in the text. In spite of the defects, 

 some of which we have tried to point out above, we have 

 no doubt the book may prove useful to telephone en- 

 gineers who are anxious to be helped over some of their 

 practical difficulties, and are not particular about a clear 

 understanding of the groundwork of their science. 



Maps: their Uses and Construction. A Short Popular 



Treatise on the Advantages and Defects of Maps on 



Various Projections, followed by an Uutline of the 



Principles involved in their Construction. By G. 



James Morrison, Memb. Inst.C.E., F.R.G.S. Pp. viii 



4- no. (London : Edward Stanford, 1901.; Price ^s. 



net. 



.\ BOOK, in English on map projections has long been 



needed, and the present work is a very welcome attempt 



to meet this need. It may be commended to all who 



have to deal with geographical questions, and to teachers 



of mathemathics and practical geometry who wish to 



find fresh exercises for their pupils. 



The volume consists of an introduction, a popular 

 account of eight common projections, followed by another 



