H 



NATURE 



\lSfov. I, 1877 



scale. The error of the determination may amount to 'i, 

 but could not, I think, exceed •2. 



I ought to add that the approximate determination of 

 the frequency of the interrupter must be made indepen- 

 dently, as the observation on the pendulum does not 

 decide which multiple of \ nearly coincides with the 

 frequency of the fork. Also the relation between the two 

 auxiliary forks was assumed, and not determined ; but as 

 to this there can be no doubt, unless it be supposed that 

 Konig's scale may be in error to the extent of a whole 

 tone. Rayleigh 



A NEW CONDENSING HYGROMETER 

 NEW apparatus of this kind, invented by M. 

 Alluard, and described by him in La Nature, is dis- 

 tinguished from all those hitherto employed by the two 



A 



Alluard's Condensing Hygrometer, 



following points : — (i) The part on which the deposit of 

 dew is to be observed is a plane well-polished face A, of 



silver or gilt brass ; (2) This plane face is set in a plate 

 of silver or brass V, itself gilt and polished, which does 

 not touch it, and which, never being cooled, always pre- 

 serves its brightness. It results from this arrangement 

 that the deposit of dew is observed with the greatest 

 facility, in such a manner that there is scarcely any differ- 

 ence between the temperatures of the instants when the 

 dew commences and ceases to appear on the instrument 

 properly cooled by the evaporation of ether. 



The form of the instrument is that of an upright prism 

 with square base. Its height is eight centimetres and the 

 side of its base eighteen millimetres. Three small copper 

 tubes pass through the upper lid ; the first reaches the 

 bottom, and the two others, one surmounted by a funnel 

 for introducing the ether, open only above. Two small 

 windows enable us to judge of the agitation of the ether 

 by the aspiration or driving back of the air intended to 

 produce coolness by the evaporation of the volatile liquid ; 

 it is best to work with an aspirator, the aspiration of 

 which we can regulate as we wish. A central tube per- 

 mits the introduction of a thermometer, /, which, placed 

 in the middle of the evaporating liquid, gives the tem- 

 perature at which the deposit of dew occurs. A small 

 sHng thermometer, fixed on the side of a brass support, 

 enables us to determine with precision the temperature of 

 the air whose hygrometric condition we wish to ascertain. 



Daniell's condensing hygrometer was formerly modified 

 by M. V. Regnault. He made it an instrument of pre- 

 cision ; but his apparatus has not been much used on 

 account of its delicate construction. The deposit of dew, 

 being made on a cylinder of polished silver, is difficult to 

 observe. In the plane face hygrometer of M. Alluard this 

 deposit is very easily seen by contrast, even at some 

 metres distance, especially if care is taken to observe in 

 such a manner as to avoid all reflection from the gilt 

 faces, when they will appear a beautiful ebony black. Its 

 employment being very simple, without losing anything of 

 its precision, there is no reason why it should not come 

 into general use. 



Since meteorological observations have multiplied on 

 all sides, the hygrometer has assumed an importance 

 which it had not before. The psychrometer is at present 

 almost exclusively employed. But all physicists know 

 that below zero we cannot trust the results which it gives ; 

 it is the same when the air is much disturbed. And yet, 

 almost everywhere, it continues to be employed on these 

 conditions. We hope that the plane face hygrometer, 

 furnished during the winter cold with an aspirator filled 

 with glycerine, will be able to yield accurate results to all 

 who do not fear to devote a few minutes to its working. 



OUR ASTRONOMICAL COLUMN 



Early Observations of the Solar Corona. — 

 Referring to Mr. Dreyer's letter in Nature (vol. xvi. 

 p. 549), the note in this column relating to the solar 

 eclipse of 1605 was by no means intended to imply that 

 it afforded one of the earliest observations of the corona, 

 nor can the eclipse of Stiklastad, as it h?s been usually 

 called, on August 31, 1030, be so characterised. Prof. 

 Julius Schmidt, of Athens, had called attention in 1870 to 

 a record of the echpse of December 22, 968, in Corfu, 

 where he found a reference to the corona, but a much 

 earlier date is assigned by Prof. Grant for the first mention 

 of this phenomenon. It occurs in Philostratus' " Life of 

 Apollonius of Tyana," Book VIII., chap, xxiii., in the 

 Leipzic edition, and runs thus : — Ilepi Se tov xpi'fw, ov iv 

 rrj 'EXXaSi fve<nTovha(,fV, iivelx^ tov ovpavop Sinarjfxia roiavrr]. 

 TOV TOV 'Hklov KVKkov TrfpifXdav aTf<f>avos, eoi/ccos "ipiSt, t^v 

 oktIvu Tjfxavpov. Prof. Grant considers that "the words 

 here quoted refer beyond all doubt to a total eclipse of 

 the sun, and that the phenomenon seen encompassing 

 the sun's disc was really as well as verbally, ideiiticalwith 

 the modem corona." He also points out that Plularch, 



