5io 



NATURE 



{March 28, 1889 



upon Neuroptera and Diptera. Among insects, the 

 Lepidoptera were studied with especial care, the type 

 selected being Thyridopteryx ephemeraiformis. The 

 careful account of this embryology, together with the 

 numerous excellent figures, entirely substantiate the 

 author's claim that the study of this type, "if it has 

 brought to light nothing new, has, in the opinion of the 

 writer at least, settled some important points connected 

 with the embryology of this group of insects." In the 

 account of segmentation and the formation of the blasto- 

 derm, the author completely confirms Bobretzky's de- 

 scriptions. The development of Thyridopteryx occupies 

 twelve quarto pages : for the details the original must 

 be consulted. The account of the embryology of Ortho- 

 ptera, represented by Mantis and the grasshopper, and 

 of the embryology of spiders, is also very complete. 



At the end of the paper many interesting and suggestive 

 conclusions are appended. Among these it is significant 

 that a writer who has done so much work upon the early 

 stages of Limulus should unhesitatingly regard this latter 

 form as an Arachnid. The Trilobites he considers as 

 " possibly the ancestral form of Limulus." 



Only a short account of this excellent paper has been 

 given here. All those interested in embryology, and the 

 light shed by it upon morphological science, will, of course, 

 make a careful study of this work. E. B. P. 



LETTERS TO THE EDITOR. 



[ The Editor does not hold himself responsible for opinions ex- 

 pressed by his correspondents . Neither can he undertake 

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

 manuscripts intended for this or any other pai-t of Nature. 

 No notice is taken of anonymous communications.^ 



The Satellite of Procyon. 



It is well-known that Procyon, like Sirius, does not travel 

 through space in a straight line, its motion undergoing changes 

 similar to those which would result from the disturbing action of 

 a large satellite. This explanation was advanced by Bessel in 

 1844, and though the satellite has not yet been seen, its exist- 

 ence has been placed beyond reasonable doubt by Dr. Auwers's 

 investigations on the subject. 



Does it not seem probable that this interesting body may be 

 revealed by the aid of photography? It is now possible to 

 photograph stars and nebulae which are beyond the visual range 

 of the most powerful tele-copes ; and if the companion of 

 Procyon, like that of Sirius, be self-luminous, there would seem 

 to be a good prospect of obtaining its image on the sensitive 

 plate. 



As the companion is doubtless in pretty close proximity to its 

 brilliant primary, it will be necessary, for photographic pur- 

 poses, to iiitercept the image of the latter by means of a suitable 

 screen. Since the direction of the satellite at any time can be 

 found from Dr. Auwers's elements, > there would be no uncertainty 

 as to the position in which this eclipsing disk (or wire) ought to 

 be placed, though its proper adjustment would be a somewhat 

 delicate operation. Should the satellite be photographed, its 

 position will become known from its configuration with respect 

 to other stars recorded on the negative. 



If a very large telescope were employed, the images of both 

 components, as distinct and separate dots, might be obtained on 

 the plate. For Procyon, a very short exposure would be requisite, 

 and this could be secured by the use of a movable stop or screen, 

 similar to that devised by Prof. Pritchard, and used for parallax 

 work at the Oxford University Observatory. 



It is scarcely necessary to dilate upon the interest which 

 would attach to a photograph showing Procyon's companion. 

 As the parallax of Piocyon has been satisfactorily determined 

 by Dr. Elkin and others — being o" '266 according to the Yale 

 College observations — we could ascertain the actual as well as 

 the relative masses of the two components. And the brightness, 

 or more strictly speaking the photographic magnitude, of the 

 satellite might also be determined with some precision. 



At present the posii ion-angle of the satellite is about 233". Its distance 

 probably amounts to but a few seconds of arc, and may be within 2". 



It may not be too late to obtain such results during the 

 present season, but exposures of four or five hours, under good 

 conditions, will not be practicable before next winter. 



J. M. Barr. 



St. Catharine's, Ontario, Canada, March 4. 



" Les Tremblements de Terre." 



The issue of Nature for February 7 (p. 337) contains a 

 review of the little work on earthquakes published for me by 

 Messrs. J. B. Bailliere. The anonymous author of the article 

 makes several criticisms on my book to which I desire to reply. 



Your critic thinks it a grave fault not to have entered into a 

 detailed description of the seismographs and seismometers at 

 present in use. He reproaches me in particular for having but 

 just mentioned Prof. Ewing's duplex pendulum seismograph ; for 

 having omitted to speak of the same inventor's horizontal pendu- 

 lum seismograph ; and especially for seeming to ignore the 

 experiments made with these instruments by Prof Ewing in 

 Japan. 1 confess that I had not been struck by the excellence 

 of the instruments in question, and that it was not through an 

 oversight that I omitted to describe Prof. Ewing's observations 

 in Japan, while I quoted in detail those of his confreres, Messrs. 

 Milne and Gray, in the same country. 



Your critic defends with some acerliity a certain class of 

 seismographs, and wrongly accuses me of failing to appreciate 

 the principles on which their construction is based. The ob- 

 jections which he makes to my treatment of M. Cavalleri's 

 pendulums of unequal length are entirely refuted from a 

 theoretical point of view by the learned note due to M. 

 Poincare, which is inserted on p. 46 of my book. I need not 

 insist further on this point. 



Your crhic thinks I have not done sufficient justice to the 

 work of the Italian savants : he forgets the limits necessarily 

 imposed on a book destined especially to give to the general 

 public an idea of the present state of an important question. 



The writer of the article regards the seismographs of to-day as 

 perfectly sufficient for all scientific needs. I am far from being the 

 only person engaged in the study of earthquakes who does not 

 share this opmion. Finally, he describes, and not very clearly» 

 the experiments which I made with M. Michel Levy to measure 

 the rate of propagation of disturbances through the soil, and the 

 registering apparatus designed for this purpose. According to 

 the writer, these experiments constitute the only advances we 

 have made in ihe study of earthquakes. If he had rendered 

 justice to our work on the subject, we should have been content, 

 and I should have raised no objections to his article. But your 

 critic reproaches us with having given results which are masked 

 by inevitable causes of error. A more careful study of the book, 

 and especially of the extracts from our original memoir, published 

 in the Comptes rendus for 1885 and 1886, would have preserved 

 hiin from so inexact an assertion. In fact, he unjustifiably mixes 

 up the preliminary experiments, made at LeCreusot by means of 

 an apparatus exactly similar to that used by Mallet and Abbott,-^ 

 with what really constitutes the basis of our work — I mean the 

 determinations made with the aid of photographic registration 

 and explosives. What your critic calls the " personal equation "" 

 of the mstrument is here nearly negligible ^ ; and the merit of 

 our method lies precisely in this point, which distinguishes us 

 from our predecessors. 



We would beg the readers of Nature to verify for themselves 

 the truth of our statement. This verification will enable then* 

 to judge of the value of the article laid before them. 



F. FouQu£. 



Finding Factors. 



It may add interest to Mr. Busk's ingenious method of 

 distinguishing between prime and composite numbers to state 

 the algebraic basis on which it rests. 



Let N be any number, and n' the next higher square number,, 

 and let N = «^ - r^, = (« -h i)^ - r-^ = {n + 2f - r^ = &c. 

 rj, Tj are formed successively from r,, by the successive additions 

 of 2« -f I, 2« -f 3, . . . the increments being in arithmetic pro- 

 gression, so that r,n = r^ + 2mn + m"-. As soon as ;-,« be- 

 comes a square, N is expressed as the difference of two squares, 

 and its factors are found. 



t See p. 219 et seg. 



^ The only subsisting cause of error is that due to the inertia of the 

 mercury, which we have estimated and introduced into all our calculations- 

 (see p. 246). 



