Appendix V. 



REPORT ON THE WORK OF THE ASTRO-PHYSICAL OBSERVATORY FOR 

 THE YEAR ENDING JUNE 30, 1895. 



During the past fiscal year the -work of the Astro-physical Observatory has con. 

 tinned to he the investigation of the infra-red solar spectrum by the bolometri*. 

 method, fully described in the report of the observatory for 1893. 



This work consists, briefly, in the production of curves, automatically traced on a 

 photographic plate by means of the bolometer aud its attendant apparatus, which 

 record by their deflections the existence, position, and intensity of (invisible) absorp- 

 tion lines in the infra-red spectrum from a prism of rock salt or from a grating. 



Such a curve, ideally perfect, would show every line in the spectrum as a deflec- 

 tion of the curve in its proper place on the plate, and there would be no deflection 

 present not due to such a solar line. A curve which fulfills the first condition as 

 regards every considerable line, is in substance already at hand, and has been 

 shown even in a former stage of the work, and it is in fact easily obtained at 

 any time by the present method, so long as we confine our attention to the prin- 

 cipal newly discovered lines; lines or deflections, that is, which are due to solar 

 and telluric causes. (These maybe called "true" lines as distinguished from the 

 smaller ''false" lines due to minute local disturbance.) I say the principal lines, 

 because those first discovered result from deflections whose amount is relatively con- 

 siderable as compared with accidental local disturbances, and in this case the true 

 lines are distinguishable (if only by their size and prominence) from the false ones, 

 the latter being due to many causes, which all, however, tend to produce more or less 

 minute deflections. This minuteness, then, fortunately, is one of the characteristics 

 of the false lines. Now, though these minor accidental deflections are negligible 

 compared with the more prominent real ones, yet as the work progresses and still 

 finer and finer details are sought, we must evidently at last reach a condition where, 

 having discovered all the larger true lines, we in seeking smaller and smaller ones, 

 finally come to such minute deflections that these, though constant, are reduced to 

 the same order of magnitude as the accidental or "false" ones, however minute the 

 latter may be. 



(In this connection it is proper to state that the plate of the spectrum showing very 

 minute detail, published in January of this year, had already been given in illustra- 

 tion of the process and printed in the Proceedings of the British Association for the 

 Advancement of Science, for August, 1894, accompanied by the statement that it was 

 presented "only in illustration," and was "not to be treated as a criterion of the 

 final results," and that the statement that all these lines had then been verified 

 escaped attention.) 



It is again fortunate, that our criterion is not that of size alone, but that in order 

 to distinguish between true deflections, however minute (due to solar and telluric 

 causes), and minute "false" deflections of the same order of size due to accidental 

 disturbances, an independent, simple, and infallible criterion exists — infallible, that 

 is, in theory, but not always easy in practical application. 



75 



