80 Contribution towards the knowledge of snakes in Brazil 
now pass on to the naming of different characteristics which 
are found between them and Lachesis insularis. 
Lachesis insularis, as it is of a more or less intense yellowish 
brown colour above; as the belly is generally of a nearly uniform 
yellowish or whitish colour; as there are no spots on the head 
nor black streak behind the eyes, and cam thus be distinguished 
at first sight from the L. jararaca and E. atroz. 
Lachesis insularis is a slender species which does not grow 
very large. The largest which I have found up to date is an L 
specimen (n.º 1.900) measuring 1 meter in length. L. jararaca, 
on ths contrary, develops much more, becoming of large diameter 
and sometimes measuring 1,400 or 1,7500, and L. atrox be- 
comes larger, measuring also 1,2500, as will be found in three 
specimens of the Butantan Institute collection (n.ºs 1.346, 2.035 
and 2036), all from the interior of the State of São Paulo. 
The tail of the Lachesis insularis is slightly prehensile whereas 
the tail of L. jararaca and L. atrox is not at all prehensile, 
The snout of Lachesis insularis is relatively narrower and 
shorter than that of L. jararaca and L. atrox; the head in the 
first mentioned is motedly wider im the temporal region than 
that of the two last mentioned species. 
Lachesis insularis constantly has 3 larger scales, more or 
less smooth, juxtaposed, as shields, which are distributed as one 
posterior pair between the. two canthals and one median anterior 
azygous, placed behind the amgle of the internasals, a disposition 
which is not found either im L. jararaca or L. atroz. 
The largest number of ventrals (195) found in L. insularis 
never attains that found sometimes in L. jararaca and L. atroz, 
Besides these principal characteristics in the anatomic order, 
there are others, in the biological order, in which L. insuluris 
is even more different from L. jararaca and L. atros. 
The characteristics of the biological order are so accentuated 
that it was through them that I was led to make the systematic 
study of the new: species. Two years after having! started obser- 
vations, with the intention of learning the biology of the Bra- 
zilian snakes, I was surprised on a certain day to find Teathers 
in the stomach and feces of 5 specimens of this species from 
a Jot which had been received from Queimada Grande Island 
on September 28th. 1919. This was an entirely new occurrence as, 
according to VITAL BRAZIL'S (29) and also in accordance with obser- 
vations made at the Butantan for many years, the Brazilian Oro- 
talinee are known(30) to feed exclusively on small rodents. 
Guided by this important indication I made a study of the 
poison which had been extracted from this species and, again 
(2) Upucia ante: 
(0) Except L. bilincata (Wied) e L. jararacussu Lacerda, which, as I verified in dissections, 
also feed on batrachians. 
