
DESTRUCTIVE DISTILLATION OF ANIMAL SUBSTANCES. 25 


Mean. Calculation. 
Carbon, . ; 26°35 26:81 Ci, 84 
Hydrogen, : 3°23 3:19 Hy, 10 
Nitrogen, . ; sg 4:49 N 14 
Chlorine, . 2 de 34:00 Cl, 106° 
Platinum, . ; 31-50 31-51 Pt 98-7 
100-00 3132 
It is clear, from these analyses, that the salt obtained is that of the base of 
which the analysis is given above; but it is equally evident, from the presence of 
small quantities of picoline, that the base itself was not obtained in a state of 
absolute purity, notwithstanding the close approximation of the experimental 
results with those required by theory. I have been struck throughout the whole 
course of the investigation by the fact, that when the fraction corresponding to 
the boiling point of any particular base has been analysed, results very nearly 
correct were obtained, even when the substance was very far from being pure. I 
found, for instance, in the earlier part of the investigation, that the fraction boil- 
ing between 270° and 280°, after one or two rectifications, gives precisely the 
results obtained from pure picoline, although on further rectification the fluid will 
begin to boil about 250°, and a small portion will still remain in the retort when 
the thermometer has risen to 300°. It is, however, readily intelligible, that this 
should be the case when we have to deal with a series of homologous bases, in 
which the per-centage of carbon goes on increasing as the boiling point rises, so 
that, as in this particular case, we have the excess of carbon in the less volatile 
base exactly counterbalancing the deficiency in the more volatile. Thus lutidine, 
containing 78°5 per cent. of carbon, and pyridine only 75:9, and each successive 
rectification removing equal quantities of the more and less volatile substances of 
which the boiling points are equidistant from that of the intermediate member of 
the series, must always leave a substance in which the quantities of the two im- 
purities must be exactly sufficient to counterbalance the error which each will 
occasion. 
Hydrargo-chloride of Lutidine.—I directed my attention to this compound, 
which is sparingly soluble and crystallizable, in hopes that it might be adapted 
to the purification of the base itself. I soon, however, abandoned it, as it turned 
out that it was not possible, in repeating its preparation, to obtain invariably the 
same substance, each base appearing, like aniline, to form different compounds with 
corrosive sublimate. When a solution of corrosive sublimate in alcohol is added to 
an alcoholic solution of lutidine, a curdy white precipitate falls immediately, unless 
the solutions be highly dilute, in which case it is slowly deposited in groups 
~ of radiated crystals. This salt dissolves in boiling water, with partial decom- 
position; it is still more soluble in spirit, and is deposited unchanged as the 
solution cools. The following analysis corresponds exactly with the formula 
— 2Hg C1+C,, HH, N. 
