136 
NATURE 
[APRIL 9, 1914 
been made of giving to the new groupings of species 
thus revealed names which are so dissimilar from that 
of the orginal genus, and from each other, as to hide 
the genus-relationship. The latter is shown when the 
genera are grouped as a family. 
The subdivision of the animal kingdom into groups 
that receive independent names should not be carried 
further than is necessary to ensure ready diagnosis of 
the species. | When carried beyond that point the 
classification is weakened. 
What is required at present is the extinction of 
probably half at least of the genera. The present 
family-group should in many cases be the genus. 
H. Cuas. WILLIAMSON, 
Marine Laboratory of the Fishery Board for 
Scotland, Aberdeen, March 23. 
The Dublin Gorilla. 
Live specimens of the gorilla are still rarities in 
British zoological gardens, and it is believed that 
except for one that has lived for several years at Stutt- 
gart, there is no example at present to be seen on the 
European continent. A few notes on a young female 
—probably about a year old—that has now lived for 
three months in the ape-house of the Royal Zoological 
Society of Ireland, in Dublin, may therefore be of 
interest to readers of NATURE. 
This litthe ape—‘‘Empress” is her name—was 
brought to Europe in company with a young male 
chimpanzee; in consequence of this companionship she 
is much tamer 
and livelier than captive gorillas 
usually are. In 
the constant 
sports which the 
two young 
creatures enjoy, 
the chimpanzee 
is*, othe aimljonme 
active and 
Spirited, tre 
quently cuffing 
the gorilla play- 
fully or dragging 
her along the 
oor WoL me 
house. The 
gorilla, however, 
is able to hold 
her own, and 
has already 
developed the 
habit of drum- 
ming on her 
chest as a _ challenge; usually she .is good- 
tempered both to her companion and to human 
visitors. She often climbs leisurely but  con- 
fidently to the top of the house. The photograph 
(by Mr. W. N. Allen) shows the little ape in a char- 
acteristic attitude, and brings out the distinctive 
shape and pose of the leg and hindquarters. Her 
eyes are very expressive, and her almost black face is 
a great contrast to the pale pink skin of her com- 
panion chimpanzee. Both the apes have completely 
recovered from an epidemic cold that ran through the: 
house in February, and it is hoped that ‘‘ Empress” 
may survive in the Dublin Gardens for several years. 
G. H. CarpENTER. 
Royal College of Science, Dublin. 
A Property of Chain-Fractions. 
For convenience, let (1; a, b, c, .) mean the 
chain-fraction of which 1/a is the first convergent, and 
a, b, c, etc., are the partial quotients. Consider all 
NO. 2319, VOL. 93] 
such fractions which have no partial quotient greater 
than g: the greatest of these is the periodic fraction 
(03,49), "and the least us (@; 9.1). We have: 
a=(153 1, 9)=(—9+ ¥ 117)/2=0-9083, 
B=(1; 9, 1)=(—9+ ¥ 117)/18=0-1009, 
and any proper fraction outside the limits (a, 8) will have 
at least one partial quotient greater than g. (The con- 
verse is not true.) More generally, one partial quotient 
at least will be greater than an assigned integer n, 
if the chain-fraction represents a quantity outside the 
interval determined by the positive roots of the 
equations :— 
o+na—n=o, nB?+nB—I1=0. 
_ As n increases, a becomes’ more and more nearly equal 
to 1, and 8 more and more nearly equal to o. The 
curious point is that if we take a proper fraction 
sufficiently near to 1 or zero, its chain-fraction expan- 
sion must contain a partial quotient greater than any 
integer assigned beforehand, and we can _ actually 
(when n is given) assign intervals containing such 
fractions and no others. For instance, when n=9 
the intervals are 
uy, (=9-- V1 17)/2 and 19, Cao /117)/18}. 
Thus 0-9089 is within the first of these intervals, and 
itsuexpansion, iS (1; iy Oy sleeyeene ae 
G. B. MaTHEWws.. 
New Units im Aerology. 
In Nature of March 19, p. 58, Prof. McAdie dis- 
cusses the question of the new units in aerology, and 
says that now is the time to agree upon a logical and 
available system, considers the megabar atmosphere 
the more appropriate, and thinks that some of the 
readers of NATURE may suggest something better. 
I have not the ambition to respond to the last sug- 
gestion, but, in order to avoid confusion in the future, 
I beg to direct attention to what has been done in 
this respect very recently. According to an official 
report, M. Pérot has presented to the French Minister 
of Commerce a report upon the reform of the legalised 
measures and weights. In this we find among the 
derived units the Newton as a unit of force 
=Kgm/sec.*, which equals 10° dynes. From this is 
derived another new unit, Pascal, as a unit of pres- 
sure, 10 Newtons per sq. cm. (10 Newtons=1 mega- 
dyne). I may add that the Calorie is proposed at 15° 
and.1-02 Pascal (=765-1 cm.). 
As France may be called the mother-country of the 
c.g.s. system, the question arises, whether the name 
Pascal might not be substituted for the modern mega- 
bar (not for ten absolute atmospheres) ? 
BoHUSLAV BRAUNER. 
Bohemian University, Prague, March 24. 
WINELAND THE GOOD.1 
RE evidence for the pre-Columbian discovery 
of North America by Norsemen depends 
essentially on two sagas: the Saga of Eric the 
Red, the Saga of Thorfinn Karlsefni in Hauks- 
book; both of which are repeated with modifica- 
tions in the Flateybook. The dates of the extant 
MSS. lie between 1300 and 1400 A.D.; the sagas 
themselves were probably composed about a cen- 
tury earlier; the main event, the discovery of 
Wineland by Leif. the Lucky, occurred. in or 
1 “Early Norse visits to North America.” By William H. Babcock. 
Smithsonian Miscellaneous Collections, vol. lix., No. 19. Pp. iv + 214, 
j x plates. (1913.) 
—_ 
