April 1 8, 1889] 



NATURE 



581 



but the explanations of the methods of determining them 

 are very meagre. Thus, on p. 22 the declination com- 

 pass is described, and the explanation given for its use is 

 simply that " the telescope is set in the plane of the geo- 

 graphical meridian, and, as the needle sets in the plane of 

 the magnetic meridian, the angle between the telescope, 

 which always lies over 0° to 180^ on the circle, and the 

 needle, is the declination." Again, referring to the deter- 

 mination of dip, it is simply stated "that when the 

 instrument is arranged with the circle in the plane of the 

 magnetic meridian, the angle of inclination can be read 

 off on the circle " (p. 26). We fear thaf'very few students 

 would succeed in getting even approximate valueswith only 

 these brief statements to guide them if the instruments 

 were put into their hands. 



One chapter is devoted to " Examples on Formula?," 

 which will, no doubt, be of great service to students, 

 although the title is rather suggestive of cramming. The 

 examples given are not less numerous than useful, no less 

 than 52 out of 144 pages being devoted to them. Several 

 of the papers set at previous Matriculation, Science and 

 Art, and other examinations are given. 



The book is well illustrated throughout, and although 

 it is more of an epitome of the chief laws and experiments 

 than a text-book, it will, no doubt, be of great service to 

 those for whom it is primarily intended. 



Heat and Light. By Edward Aveling, D.Sc. (London: 

 Chapman and Hall, 1889.) 



This is another text-book of the series referred to in the 

 preceding notice, and follows on the same lines. It is 

 characterized by the same bare outline, the expl mations 

 of the methods of determining the various data generally 

 including no suggestion whatever as to difficulties and 

 corrections. This is especially noticeable in the account 

 of Joule's classical experiment (p. 26), in which no men- 

 tion whatever is made of the corrections for loss of heat 

 due to radiation or for the velocity of the weight on 

 falling. The diagram, too, is seriously wrong, since it 

 simply shows a set of vanes revolving in a vessel of 

 water ; without the pierced partitions necessary to pre- 

 vent the rotation of the water, the experiment is, of 

 course, useless. 



The chapter on the composition of white light and the 

 spectrum is perhaps the least satisfactory in the book. 

 The merest outline of the subject is given, and there are 

 two or three very obvious slips. On p. 165, for instance, 

 the electric light and the lime-light are quoted as examples 

 of monochromatic light, and again on p. 166 it is stated 

 that "glowing gases yield spectra with dark lines." The 

 idea that the actinic rays are confined to the violet part 

 of the spectrum is rather old-fashioned, and is scarcely 

 •likely to be credited by a student who may have hap- 

 pened to experiment in the direction of orthochromatic 

 photography. 



Like its predecessor, the book contains numerous 

 examples and illustrations. 



The Encydopcedia Briiannica. Ninth Edition. Index. 

 (Edinbuigh: Adam and Charles Black, 1889.) 



The publication of this volume (of 500 pages) marks the 

 completion of one of the greatest literary undertakings 

 of the present age. As to the necessity for an index 

 there can be no doubt, since, as the editor explains, the 

 plan of the " Encyclopaedia Britannica" was that subjects 

 rather than words should be dedt with, and that large 

 subjects should be discussed in a connected way, under 

 general headings, so that the book might be used not only 

 for occasional reference, but for systematic study. This 

 plan was adhered to, and the result is that " many things 

 which a reader may wish to understand are explained, 

 not under their own names, but in the course of a larger 

 discussion." In such cases reference must be made to 

 the index ; and this is so full and so accurate tha: no one 



who may have occasion to consult it will ever have the 

 slightest difficulty in at once finding what he wants. The 

 index has been compiled by Mr. William Cairns, and 

 arranged and revised by the Rev. George M'Arthur, with 

 the assistance of Miss Emily Stevenson and Mr. J. T. 

 Bealby. The volume contains also a complete list of 

 contributors, with a key to the initial letters affixed to the 

 longer articles. A glance over this list, which includes 

 almost all the foremost writers of the day, suffices to 

 explain the high character of the work as a whole. 



Blackie's Modern Cydobcedia of Universal Information. 



Edited by Charles Annandale, M.A., LL.D. Vol. I. 



(London : Blackie and Son, 1889.) 

 It is intended that the work of which this is the opening 

 volume shall serve as " a convenient work of reference 

 for readers of all classes— comprehensive in scope, handy 

 in size, moderate in price, and generally adapted to the 

 needs of the day." Of course no one who may want to 

 obtain a thorough knowledge of any subject will think of 

 seeking for it in such a work as this ; but the editor does 

 not place before himself too high an object of ambition 

 when he expresses a hope that the new Cyclopaedia may 

 prove useful to persons who have little time for acquiring 

 information from books in general, though they take an 

 interest in many topics lying outside their own pursuits. 

 The present volume deals with words beginning with the 

 letter A, and with many of those beginning with B. The 

 articles are short but clear, and, so far as they go, 

 accurate. Especial attention has been given to matters 

 which are of living interest in our own day, and we are 

 glad to see that many scientific articles have been written 

 or revised by specialists. The volume contains some 

 good maps and many interesting pictorial illustrations. 



LETTERS TO THE EDITOR. 



[ 77^1? Editor does not hold himself responsible Jor opinions ex- 

 pressed by his correspondents . Neither can he undertake 

 to return, or to correspond zvith the writers of, rejected 

 manuscripts intended for this or any other part of NATtJRE. 

 No notice is taken of anonymous communications. ^ 



Spherical Eggs. 

 A BIOLOGICAL friend lately asked nie for a solution of the 

 prohL*m, How many spherical ej^gs, 003 millimetre in diameter, 

 can be contained in a cubic millimetre?, the whole space occu- 

 pied by the eggs being lar^e compared with a millimetre. 

 Assuming the eg^s as closely packed as po>sible in a horizontal 

 stratum, their centres will lie at the angular points of a series of 

 equilateral triangles whose sides are equal to a diameter. The 

 number of spheres in this stratum corresponding to a unit of 



area will thus, on an average, be ,,, , </ being the diameter. 



The next stratum above will have the centres of the spheres 

 placed so that each is at the vertex of a regular tetrahedron. 



having alternate equilateral triangles of the lower series as base. 

 Evidently, in a volume of the slice between the two planes of 

 centres, having a unit of area for its base, there will be, on 



an average, — ? ^ spheres. But the thickness of the slice is 



Hence, in a unit volume, on an average, there will be 



v 



n/2 



spheres, i.e. sj2 times, or about |, as many as would be 



contained, assuming their centres at the corners of cubes. 



