The Ultra-Violet Absorption Spectra of Blood Sera. 331 



on the other side of the band comprises two sections ; section & describes a 

 short, nearly vertical fall from the head at wave-length about 2800, and 

 an abrupt projection towards the ultra-violet to a point where the curve 

 turns sharply downwards again. Section e extends from this point to the 

 depression at wave-length about 2540, whence section £ rises steeply to a 

 high value at a wave-length about 2400. A further section, 77, has been 

 observed with three or four pathological specimens, and is indicated in the 

 figure by a dotted line. 



The peculiarities of the central band call for special attention. Section 7 

 exhibits slight irregularities of form which are found to be constant. Among 

 these, the most evident are slight step-like prominences at about a quarter 

 and half-way down from the top, and one very near the bottom, which 

 is often well developed. Section 8 includes with the upper part of 

 section 7 a small area of elongated shape, which varies somewhat in altitude. 

 Experience shows that the section e is the part most subject to variation 

 (vide infra). Section 7 seldom varies much except when, as in some 

 pathological sera, the greater changes in 8 or e extend their influence across 

 the band. Section /3 is subject to some variation ; section £ has rarely 

 been found to alter materially. 



Between sixty and seventy specimens of normal blood have been examined, 

 with results which are practically constant. The general character of the 

 curve has never altered. In the course of their study, the various 

 peculiarities of the absorption spectrum have become recognised gradually, 

 and the method of procedure improved from time to time so as to develop 

 the several parts of the curve. The chief conclusions arrived at are 

 recorded in the description already given. 



Sex and age have not so far found any well defined expression in the 

 properties of the absorption curve. On the other hand, it cannot be said 

 that there is no general differentiation. The central band appears to be 

 slightly narrower in the female, and to be shallower in the child, but one 

 merges into another so gradually that nothing more precise can be said 

 until a considerable number of specimens have been very carefully com- 

 pared under the best conditions, taking full advantage of all earlier 

 experience. 



Inasmuch as serum is a solution of a mixture of- substances in somewhat 

 variable proportions, smooth absorption curves comparable with those 

 obtainable with pure substances cannot be expected. The serum curve is 

 rather the resultant of several superposed curves, each characteristic of 

 some constituent of the fluid. It becomes of interest, therefore, to 

 separate these members and to determine their individual absorptions with 



