6 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 97 
and they may continue to be known as splendidus (magdalenae). 
Additional details concerning the above are given under the appro- 
priate subspecies headings. 
Further study of the variation and distribution of Sciwrus granatensis 
reveals that other forms, previously considered distinct specifically, 
grade into it. These are listed and discussed separately. The status 
of S. pyrrhinus Thomas is obscure. Specimens in the United States 
National Museum from Chanchamayo and La Merced, Peru, agree 
with the description of pyrrhinus externally, but their cranial measure- 
ments differ widely and clearly show that they are members of the 
Hadroscwurus group. The author’s notes on the type merely indicate 
that it is a large squirrel, probably of the Hadrosciurus group. This 
species has been considered by Allen (1915, p. 254) a form of his genus 
Mesoscwurus most nearly related to saltuensis (=granatensis). Other 
species generally included in the “ Mesosciurus’”’ group are richmondi 
and argentinus. ‘These are not dealt with here and may be distinct 
from granatensis. 
The following additional named forms, listed and discussed in 
chronological order, are considered synonyms of granatensis. Most of 
these names are retained as valid subspecifically. 
Scirus chrysurus Pucheran (1845, p. 337). This is an immature of 
Sciurus hyporrhodus Gray (1867), which name it replaces as it has 
been described from the same locality, ‘‘Santa-Fé de Bogoté.” The 
type matches perfectly with very small, immature representatives of 
granatensis from the Bogoté region. It has been confused by revisers 
with the very different S. rufoniger described from the same locality 
by the same author (1845, p. 336). 
Sciurus gerrardi Gray (1861, p. 92) described from ‘‘ New Grenada.” 
The use here of the earlier name granatensis for the common squirrels 
of Colombia eliminates gerrardi as a species. Alston (1878, p. 666) 
had examined the types of both gerrardi and variabilis and declared 
them to be exactly synonymous. Thomas (in Allen, 1915, p. 240, 
footnote 1) concurred. By suppressing the name variabilis, Allen 
was compelled to resurrect gerrardi as the name for the common red 
squirrel. After considerable difficulty he decided (1915, p. 308) to 
restrict the type locality of gerrardi to somewhere between the range 
of the “ zuliae-cucutae group” and ‘‘baudensis’”’; a specimen from the 
Rio San Jorge was considered typical. This must be accepted, and 
the squirrels of that region may be known as S. granatensis gerrardi 
and distinguishable from the typical form by the redder color and the 
black-tipped tail, if this last is, indeed, a valid distinction. 
Sciurus hoffmanni Peters (1863, p. 654). Examination of specimens 
from Costa Rica, Panama, Colombia, and Ecuador representing all 
the forms recognized by Allen (1915) as races of hoffmanni and com- 
parison of these with the material which forms the principal subject of 
