134 REPORT— 1903. 



Experimental Details. 



For the specimens of the opium alkaloids, including gnoscopine, we 

 are indebted to the kindness of Messrs. T. and H. Smith, of Edinburgh, 

 and for the specimens of oxyhydrastinine and w-aminoethylpiperonyl- 

 carboxylic anhydride, to Professor AV, H. Perkin, jun. The specimens 

 of inactive and artificial corydaline, corybulbine, tetrahydroberberine, 

 tetrahydropapaverine, dehydrocorydaline, corydic and berberidic acids, 

 corydaldine, and hydrastinine were prepared in the laboratory of the 

 University College of North Wales, Bangor ; and our best thanks are 

 due to Messrs. C K. Tinkler, K. S. Caldwell, and Ed. Jones for assist- 

 ing in the preparation of some of these substances, and to Mr. C. P. 

 Finn for assisting in photographing the spectra. The remaining alka- 

 loids were obtained by purchase. In every case the specimens were 

 tested as to their purity, and, where necessary, subjected to purifica- 

 tion. Whenever possible, specimens were obtained from at least two 

 distinct sources, and several independent examinations were made of each 

 specimen. 



In photographing the spectra and in representing them graphically, 

 owing to the slight solubility of some of the substances examined, it was 

 not always possible to get a solution of 1/100, and thicker layers of a 

 more dilute solution had to be employed. In such cases, for convenience 

 of reference, 25 mm. of a solution of 1/500 have been plotted as equi- 

 valent to 5 mm. of a solution of 1/100. Except in the case of hydrastine, 

 all the curves are drawn to the same scale. The position of the trans- 

 mitted portions of the spectra and of the absorption band have been 

 marked on one of the curves (fig. 3). 



We may remark that it is difficult by means of curves to give a proper 

 representation of the photographs, it being found impracticable to 

 adequately represent differences of intensity as well as extent of absorp- 

 tion upon which the similarity or difference between two series of spectra 

 often to a large extent depends. 



By far the most satisfactory comparison is that which is made by the 

 actual inspection of the photographs. When reasonable care is taken 

 to work under similar conditions, the results are remarkably constant. 

 We have never discovered any discrepancy between the photographs of 

 the same preparation, even when taken by different operators at wide 

 intervals of time. Hartley having worked in 1882 and again in 1884 

 with constant weights and not with molecular quantities of the alkaloids, 

 it was found necessary, for purposes of comparison, to repeat the exami- 

 nation of a few of the alkaloids which he had previously examined. In 

 so far as it is possible in such cases to compare the results, they show 

 remarkably close agreement. 



