April 28, 1898] 



NATURE 



621 



As at first constructed, the strips, representing arms of the 

 VVheatstone bridge, were made of iron from oooi to O'oooi of 

 an inch in thickness. The instrument was, even under these 

 initial conditions, very many times as sensitive as the best ther- 

 mopile the writer then possessed, but there does not appear to be 

 any definite statement as to the exact sensitiveness in its early 

 form. 



In the article referred to, however, the instrument is repre- 

 sented as giving a deflection of about 40 scale divisions (milli- 

 metres) from the lunar heat, concentrated by a thirteen-inch 

 lens, and it was sufficiently accurate to give a probable error of 

 rather less than one per cent, for a single observation on a con- 

 stant source of heat, so that the accuracy of the bolometer 

 (quite a distinct consideration from its sensitiveness) was even 

 then as great as that of the best photometric process. The 

 galvanometer in use at that time was one of the early Thomson 

 pattern made by Elliott. 



The first bolometers were made by the writer's own hands. 

 Subsequently the strips were usually cut out from sheets of thin 

 platinum, and in one or two instances made from flatted wire, 

 the strip of the linear bolometer at that time (about 1883) being 

 usually about 10 millimetres long ; anywhere from 0001 too'Oi 

 of a millimetre thick, and, according to its special purpose, 

 being made from i millimetre to 01 millimetre wide. 



About 1886 the mounting of the instrument had been 

 improved by the writer, so that the strip appeared like the 

 ; vertical "wire" of a reticule in the focus of a positive eye- 

 piece. It was also movable in some cases by a micrometer 

 screw, and was, in fact, a micrometer thread controlled in the 

 usual way, but endowed with the special power of feeling the 

 radiations from any object on which it was directed. 



In the earliest spectrum work the bolometer developed 

 another important quality, its "precision." This quality is 

 quite independent of the accuracy with which it repeats 

 measures of radiation or any constant source of heat, and 

 concerns the precision of setting, as a micrometer 1 bread. It 

 could, even twelve years ago, be pointed, not only like the ther- 

 mopile, within a fraction of a degree of the place of the source 

 of radiation, as for instance on a bright line in the spectrum, but 

 within a fraction of a minute of arc. 



The instrument of course depends for its general efficiency on 

 the gahanometer with which it is connected. That used in 

 1886 ' had several improvements due to the suggesti<ms of Sir 

 William Thomson and Prof. Rowland, and was perhaps, at that 

 time, the most effective instrument of its kind in use for such a 

 purpose, the mirror and needles having been specially con- 

 structed at the Allegheny Observatory. The mirrors were 

 platinised by the kindness of Prof. Wright, and were at that 

 time nearly a centimetre in diameter. The needles were hollow 

 magnets made by Mr. Very of the Allegheny Observatory. 

 For the damping mechanism of the older galvanometer, I had 

 substituted a dragon-fly {Libelltila) wing, in which nature off"ers 

 a model of lightness and rigidity quite inimitable by art. At 

 that time, when making a single vibration in 20 seconds, a 

 deflection of one millimetre division of the scale at one metre 

 distance was given by a current of ooc)0,cxx),ooo5 amperes, the 

 instrument as described being capable of recording a change of 

 temperature in the bolometer strips of less than o'ooooi of a 

 degree Centigrade. So much less than this could be observed 

 by special precaution, that it might be said that this one one- 

 hundred-thousandth of a degree was not only indicable but 

 measurable by the apparatus, which was employed as described, 

 in the determinations of the relations of n to A for the rock-salt 

 prism, and by which the infra-red spectrum was at that time 

 followed by actual measurement, to a wave-length of rather over 

 five one-thousandths of a millimetre. 



Since then the bolometer has been used in various researches, 

 of which some occasional account has been given in the American 

 Journal of Science. (See numbers for November 1888, and August 

 1890.) During recent years it has been specially employed in 

 rnaking a holographic map of the lower spectrum, the publica- 

 tion of which has been greatly delayed by conditions incidental 

 to the relations of the Smithsonian Observatory with the 

 Government, but which it is hoped will not be deferred much 

 longer. 



Without here entering into an account of the work done by it, 

 I have thought that it might be of interest to give very briefly a 

 statement of the present condition and form of the instrument 

 itself, considered under three aspects. 



1 American Journal of Science, third series, vol. xxxii. p. 90, 1886. 



NO. 1487, VOL. 57] 



(i) Its precision, or the degree of exactitude with which it can 

 be set on a special point, as, for instance, on a line of the in- 

 visible spectrum, recognised by its heat radiation alone. 



(2) Its accuracy, or its capacity for repeating the same 

 measure of radiation under like conditions. 



(3) Its sensitiveness, or capacity for detecting minute radia- 

 tions. 



The instrument which I will take as the subject of comparison 

 with the earlier one as described in this Journal (August 1886), 

 is now in use in a chamber automatically kept at a temperature 

 constant within one-tenth of a degree Centigrade. 



The strips, the essential part of the instrument, are in the 

 present case made by Mr. C. G. Abbot, and are of platinum, the 

 central one being rather less than o"i of a millimetre wide. 

 (The case is now made of metal instead of ebonite, and is 

 surrounded by a current of water). 



It is quite possible to make bolometer strips much narrower, 

 but this is less necessary with the employment of the long-focus, 

 image-forming mirror, so that in the present case the strip is at 

 such a distance that it subtends an angle of 3 '4 seconds. Its 

 angular aperture is in practice adapted to that of the slit, which, 

 with the use of the long collimator employed by the writer, 

 gives a capacity of pointing (pointing, that is, in the dark), with 

 a probable error of little over a second of arc. Quite recently, 

 owing to the use of a novel collimating system of two cylindrical 

 mirrors proposed by Mr. Abbot, the slit, though at a moderate 

 distance, can have an opening sufficient to avoid prejudicial 

 diff^raction effects, while subtending an angle of considerably less 

 than one second of arc. 



In the galvanometer, the use of the fine quartz threads and 

 specially small mirrors, originally due to Mr. Boys, has lately 

 been carried to what seems near the practicable extreme, the 

 quite invisible thread being made some 30 centimetres long, the 

 mirror 2 millimetres in diameter, and weighing but 2 milli- 

 grams, and its six needles, of proportionate weight and dimen- 

 sions. 



This system is now made to serve with a much shorter swing 

 than that formerly employed. If we reduce it to a time of 

 single vibration of 20 seconds, only for the purpose of comparing 

 it with the values already given in the earlier form, we obtain 

 the results submitted below. 



Before giving them, however, it is to be mentioned that the 

 apparatus at VVashington is most unfavourably situated, owing 

 to its being subject to tremor from the traffic of neighbouring 

 streets and to other causes, which it has been the object of 

 years of struggle to conquer. This has been so far done that 

 the values presently to be given (which, it will be remembered, 

 are only attainable in a chamber of constant temperature, with 

 special precaution against disturbance from external tremor), 

 can be counted on as real values, always obtainable under proper 

 conditions, and, in fact, rather within than without the average 

 working capacity of the instrument. 



I here consider the bolometer as at present employed. 



(1) With regard to its precision, or exactness of pointing. 

 The old thermopile could be set on a portion of the spectrum 

 only vvith an error of a considerable fraction of a degree. The 

 linear bolometer as employed in 1886 could be set with a prob- 

 able error of a fraction of a minute of arc. The bolometer as 

 employed to-day, and moved through the spectrum by clock- 

 work, can be automatically set with a probable error of a single 

 observation of little over a second of arc, can be set, that is, in 

 the dark with a precision little inferior to the capacity of the 

 eye in setting a micrometer thread in the light. 



(2) As to its accuracy. I have had occasion recently to take a 

 series of measures of successive throws of the galvanometer, 

 using as a source of heat an Argand petroleum flame in a 

 common student's lamp. I had no photometer at hand, but 

 taking the usual statements of the text-books as to the accuracy 

 of vision, it might be expected that such measures with the eye 

 would give a probable error of about one per cent. (This is 

 where sources of light of similar quality are compared. ) The 

 probable error of a single galvanometer reading was between 

 0*03 and 004 of one per cent., and this included the fluctuation 

 of the intensity of the source of radiation, and the error of 

 estimating tenths by the reader on the scale, both quantities of 

 nearly the same order as the error in question. It seems safe to 

 say, then, that no error attributable to inaccuracy of the bolo- 

 meter could be detected by the means employed. 



(3) As to sensitiveness. In the early work, for a time of 

 single swing of 20 seconds, a deflection of one millimetre with 



