126 



NA TURE 



June i i, 1S96 



and which appeared to have an enormous internal resistance, 

 though its blue appearance and other indications pointed to 

 rather a low vacuum, which seems to show that this is the case. 

 A. A. C. SWINTON. 

 66 Victoria Street, S.W., June 8. 



Dalton's Atomic Theory. 



With reference to the communications from the authors and 

 from the reviewer of the " New View of the Origin of Dalton's 

 Atomic Theory," published in Nature for May 14,'! beg 

 leave to offer the following remarks. The most serious difficulty 

 which the reviewer advances against the new view, seems to be 

 that Dalton, in his manuscript lecture to the Royal Institution 

 in 1810, states that, as a consequence of an idea respecting 

 elastic fluids which occurred to him in 1805, "it became an 

 object to determine the relative sizes and wciglils, together with 

 the relative number of atoms in a given volume " ; whereas in 

 one of his note-books, under date September 6, 1803, a table of 

 atomic weights is given. The reviewer says : — " The authors 

 notice this conflict of statement, but get rid of it by assuming 

 1805 to be a clerical error for 1803." In regard to these 

 conflicting dates, I beg to draw attention to a passage which 

 appears to have escaped the vigilance both of the authors and of 

 the reviewer, and which seems to tell strongly in favour of the 

 clerical error theory. In the preface to Part I. of Dalton's 

 "New System of Chemical Philosophy" (1808), the author, 

 writing of himself, says : — "In 1803, he was gradually led to 

 those primary laws, which seem to obtain in regard to heat, and 

 to chemical combinations, and which it is the object of the present 

 work to exhibit and elucidate. A brief outline of them was first 

 publicly given the ensuing winter in a course of lectures on 

 natural philosophy, at the Royal Institution in London, and 

 was left for publication in the journals of the Institution ; but 

 he is not informed whether that was done." I do not think 

 there is any room for reasonable doubt that this passage refers, 

 amongst other things, to the same idea as that stated in the 

 manuscript lecture to have occurred to Dalton in 1805. In any 

 case the date 1803 is definitely settled by the .sentence referring 

 to the lectures at the Royal Institution, since we know that 

 Dalton's lectures were begun there on December 22, 1S03 

 (compare Roscoe and Harden's " New View, &c.," p. 61). It 

 ought to be possible to place this matter Ireyond all doubt if 

 the notes stated by Dalton to have been left for publication in 

 the journals of the Royal Institution are forthcoming. 



Lf.o.nard Dobbin. 



University of Edinburgh, May 15. 



Halley's Chart of Magnetic Declinations. 



I AM again able to add another reference to the list of 

 publications of Halley's Chart of Magnetic Declinations (see 

 Nature, vol. Hi. pp. 79, 106, 343). 



The chart to which I now refer is one of the plates of Peter 

 van Musschenbroek's work, entitled " Phy.sicae Experimentales 

 et Geometricae de Magnete, Tuborum Capillarium Vitre- 

 orumque Speculorum Attractione, Magnitudine Terrae, Coliaer- 

 entia Corporum Firmorum " ; Lugundi Batavorum, Miiccxxix. 

 Its size is igj inches ;•; 7^ inches, and it takes in the entire 

 circumference of the globe. The title, in the upper left-hand 

 corner, reads : " Tabula Totius (Jrbis Terrarum Exhibens 

 Declinationes Magneticas, ad Annum 1700. Composita ab 

 Edmundo Halleyo. Simul eum Inclinationibus a Poundio 

 Observatis." Chas. L. Clarke. 



New York, May 28. 



Professional Qualifications. 



I am anxious to prepare myself for the appointment of 

 professor or teacher in chemistry at one of the new technical 

 schools held under the County Councils. Will you kindly 

 inform me the best way to become competent for the post ? 

 My age is twenty-five, and I hold first-class certificates in 

 advanced chemistry at South Kensington Science and Art 

 examinations. Is it necessary to obtain the F.I.C. or some 

 similar degree first ? Any hints you could give me would be of 

 great help to me. 



I must add that at present I have had no experience in 

 teaching. Stument. 



NO. 1389, VOL. 54] 



LEAP-YEARS AND THEIR OCCASIONAL 

 OMISSION. 



A FTEK the present year there will be no leap-year, 

 ■'*■ at any rate, in the many countries which now 

 observe the Gregorian style, until 1904 ; in other words 

 1900, which would, by the Julian rule, have been a leap- 

 year, will be a common year and have to content itself, 

 like the three years preceding and the three years follow- 

 ing it, with the ordinary number of three hundred and 

 sixty-five days. Only once has a similar omission 

 occurred before since the reformation of the calendar in 

 England, viz. in 1800, a year remarkable enough in other 

 respects. The change was originally made in 1582 ; but 

 as centuries di\isible by four hundred without remainder 

 were to be considered leap or bissextile years by either 

 reckoning, there was only occasion, in 1700, when a year 

 was observed as such in England, which was a common 

 year in southern Europe ; for 1600 was, as 2000 will be, 

 a leap-year by the Gregorian as well as by the Julian 

 reckoning. Few persons seem to recollect that the 

 change which was effected at Rome in 1582, and followed 

 in this country in 1752, was twofold in its character. If 

 it be desired to make the date in any year correspond 

 exactly with the season of the year, this can of course be 

 done for any future time by inserting or omitting certain 

 intercalary days in the calendar in some such way as is 

 directed by the Gregorian rule to which we are now 

 accustomed, and which was de\ised by Clavius under 

 the authority of Pope Gregory XIII. But if this had 

 not been done in past ages through want of exact know- 

 ledge of the true length of the year, or from any other 

 cause, the fact may either be accepted as inevitable and 

 therefore regretfully disregarded, or we may, if we wish, 

 so change the existing dates in the year from which we 

 start, as to make the seasons correspond with what they 

 were on these dates at some definite period in the past. 

 This is what was actually done, the period selected being 

 A.D. 325, the year of the first great Council of the Church 

 held at Nic;ta in the reign of Constantine the Great. 

 Atthat time the vernal equinox fell on .March 21 ; and as, 

 in consequence of the observance of the Julian length of 

 the year in the interim, it fell in 1582 on the nth of that 

 month, it was decreed that in the following autumn ten 

 days should be struck out of the calendar, by calling the 

 day after October 4 the 15th, so that in future the 

 vernal equinox (and all the other seasons) should fall 

 as they had done in 325. This arrangement involved 

 another inconvenience besides the awkward enumeration 

 of days in that year, viz. that the seasons were made to 

 disagree appreciably with their dates in the years and 

 centuries immediately preceding the time of the change. 

 However, on the whole, it was thought to be the best 

 arrangement, and it was gradually followed by most of 

 the nations of Europe excepting Russia. In England 

 the change was made in 1752, and the calendar in all 

 respects assimilated to that of the New Style, adopting 

 the Gregorian rules. As in accordance with these, 

 1700 had not been a leap-year, whereas in England by 

 the Julian reckoning it had been, the two calendars now 

 differed by eleven days ; the -Act of Parliament therefore, 

 which ordered the change, enacted that the day after 

 September 2, 1752, should be called the 14th. 



In speaking of the erroneous length of the year 

 assumed in the Julian calendar, we used the expression 

 "through want of knowledge of the true length of the 

 year, or from any other cause." This was intended as a 

 reference to the fact that, although the exact length of 

 the year was not known in the time of Julius Ciesar, it 

 was certainly known that it fell several minutes short of 

 365^ days. But it seems that he thought this was 

 sufficiently near for all practical purposes ; and a dis- 

 tinguished American astronomer of our own day, in the 

 light of all our modern improved knowledge, is of that 



