2 
Proceedings of the Biological Societi/ of Wasiiingtoo. 
Sjiite of other evidence whiclt seetns conclusive that it was. Tliis evidence 
need not be detailed, ])ut it includes the ori.cjinal i)rospectns, the admitted 
regnlaritj' for the first thirteen volumes, the fact that the scattered dates 
of the later volumes occur on the parts that would in regular course have 
been issued on those dates,* the fact that sncb scattered dates occur even 
in the next to the last volume practically at the end of the series,t and 
the correspondence of the issue of the last part with the date of Doctoi- 
iShaw'’s death wdiicli occurred on July 22, ISld. The last volume, there¬ 
fore, closes with the part for July instead of that for August as otherwise 
wonld have been tbe case. In the face of this evidence of regularity, the 
mere absence of printed dates even on a large nnndjer of the i)arts has 
scarcely any weight. The number of plates per part varied from tbree 
to four, but, as first noticed by Doctor Richmond, practically ever}^ part 
begins with a bird and thus it is simple to distinguish parts in the bound 
volumes when they are not otberw’ise ol)vions. Knowing the date of 
beginning and of ending, the number of parts, and tbe interval between 
issues, tbe determination of the date of any particular part is a matter of 
very simple mathematics and exceedingly conclusive. By this method, 
Ovis canadensis, wdiich is in an undated part, is shown to have appeared 
in February, 1804,1: and the presumption tliat it appeared February first 
is so strong as to be accepted for ])ractical purposes. 
Ovis cervina Desmarest. 
This name was first published in the twenty-fourth and last volume of 
the “Nouveau Dictionnaire d’llistoire Natnrelle.” This volume and the 
two preceding ones, issued simultaneously, bear the title page date 1804, 
while volumes I-XXI are dated 1803, although volumes I-III, as appears 
below, were actually issued in 1802. The series was issued quarterly in 
“ livraisons ” of three volumes each and the exact dates of their i)nblica- 
tion are indicated by regular notices winch ap])eared in the “Journal 
Typographique.”§ The twenty-fourth volume is noticed in the issue for 
March 7, 1804, and that therefore is the approximate date of the name 
Ovis cervina. 
Since tliey will doubtless be of use in other comiections, the notices of 
the entire series are given, as follows: 
*In the very few cases where there is not absolute affreeinent, tin- date on the plate 
is always earlier, never later, than the contingent date of issue. 
t This evidently escaped Doctor Allen, possibly on account of imperfection in the 
copy of tlie work examined by him. In the Kidd Museum copy and the Tabrary of 
(’ongress copy, Plato 1004 of Vol. XXIII is dated 1.S12. In this connection, it may be 
stated that these copies directly disagree with Doctor Allen's statements in tlie following 
particulars: Vol. II, Pis. 44, (V_>, 09, and 72 are dated; Volume III, PI. 7,s is dated; Vol 
IV, Pi. 124 is dated; Vol. V, I’l. 152 is not dated; Vol. XXIII, Pi. 1004 is dated. 
jThis coincides with Doctor Allen’s conclusion that" careful collation of Shaw and 
Nodder’s work given below shows that the actual date of publication of this name was 
almost umtuestionaldy I'Mbruary, 1S04.” (1. c., p. 11.) 
§“A weekly .journal, published in Paris from 1797 to 1810, devoted exclusively to the 
announcement of now works, music, etc.; succeeded in ISIl Ijy the ‘.lonrnal General de 
PImprimerieetde laLibrarie,’ ” Richmond, The Auk, XVI, p.:J26, footnote, October 1899 
