320 
Already, however, in history for college stu- 
dents—witness Prof. Bury’s recent single volume 
upon Greece—there is a tendency to keep in view 
the philosophy of the subject as a higher stratum 
based upon the facts; and something of the kind 
may be possible in our text-books of science as 
time goes on. At present one would like to re- 
commend a pupil to read no text-book, but to buy 
a number of the shilling volumes written by 
specialists on the lines and subjects that attract 
themselves. This, however, would not enable 
the student to meet the requirements of a uni- 
versity degree. Prof. Park is well aware of this, 
and has kept himself within traditional bounds. 
At the same time he makes good use of recent 
work, including even the Piltdown skull (p. 480) ; 
and his references to New Zealand and the 
southern hemisphere introduce a welcome series 
of examples. The illustrations are numerous and 
well chosen, though those of ammonites lead to 
the retention of a somewhat old classification. 
Formule might have been more freely utilised to 
show the composition of the  rock-forming 
minerals, which are here rather loosely described. 
The absolutely, essential boron would then have 
appeared: as a constituent of tourmaline, and the 
rhombic pyroxenes would not have been defined 
merely as “variable silicates.” ‘‘Titanite” (pp. 
197 and 108) is not a synonym for the titanic iron- 
ores. The explanation of technical terms founded 
on Greek words ‘is a very useful feature through- 
out the book, | ** GAAS 
The West India Committee Map of the West 
Indies..” Seale 55 miles=1 in. Size 3 ft. 9 in. by 
2 ft. roin. (London: George Philip and Son, 
Ltd., 1914.) Mounted on cloth and varnished 
with rollers, ros. 6d.; on sheet, unmounted, 
So) OGL: 
/ 
Tue object of the West India Committee in issuing 
this» map‘isto stimulate interest in the British 
West ‘Indies.” No attempt has been made to 
give the land relief and other physiographical 
features of the individual islands—the relative size 
of the islands in relation to the parts of the main- 
land shown making such a course impracticable. 
Prominence given to railways, sea routes, 
cables, and other data of commercial importance. 
Three inset maps are provided: one shows the 
routes and distances between Canada and _ the 
United States and the West Indies, another a plan 
of the Panama Canal, and a third a graphic repre- 
sentation of the.areas and populations of the 
islands. 
The Origin of the World. 
By R. McMillan. Pp. xiii+136. (London: 
Watts and Co., 1914.) Price 2s. net. 
THE object of this little book is to explain, in lan- 
guage simple enough for an intelligent child to 
understand, the steps in the evolution of our 
planet, of plant and animal life, and of the human 
race. The book is written in a pleasant style 
which should appeal to young readers, arrest their 
attention, and engender a desire for fuller know- 
ledge. 
NO. 2326, VOL. .93]| 
1S 
A Book for Children. 
NATURE 
[May 28, 1914 
LETTERS TO THE: EDITOR. 
[The Editor does not hold himself responsible for 
opinions expressed by his correspondents. Neither 
can he undertake to return, or to correspond with 
the writers of, rejected manuscripts intended for 
this or any other part of Nature. No notice ts 
taken of anonymous communications. | 
Temperature-Difference between the Un and Down 
Traces of Sounding-Balloon Diagrams. 
Dr. VAN BEMMELEN’S letter in Nature of May 14 is 
| - 
/ 
of great interest to me, and seems to prove an appre- 
ciable amount of lag in the instruments he uses. 
Every thermometer must as a matter of course have 
a certain amount of lag, but I have not been able in 
the records of the English instruments to detect any 
sign of it, although there is a marked distinction 
between records obtained at night or when the sun is 
low, and those obtained when it is high. 
In the diagram, A is the ordinary type of a night 
ascent, B that of a day ascent. The double trace, one 
made on the ascent, the other on the descent, is 
apparent in about every record obtained. It is not 
often apparent to the naked eye, and hence the dia- 
gram is an exaggerated one, but under the micro- 
scope by means of which the records are read, the 
traces in type A can be seen to cross each other here 
| B 
and there, in tvpe B, on the other hand, the traces 
are distinct throughout, but the distance between them 
is plainly variable, ranging often from about 1° to about 
3- C. It is very seldom that differences so large as 
4° C. are found. In three cases out of four type A 
will occur at night, and type B when the sun’s aiti- 
tude exceeds 10°, but now and then type B occurs at 
| night, and seems then to indicate an actual change of 
| temperature during the ascent. 
| In the English instruments the thermograph de- 
| pends on the temperature of a very thin strip of 
German silver; this is kept stretched by a small invar 
tube. The expansion of the invar is nil, and therefore 
its temperature is of no consequence; the German 
silver is 0:03 mm. thick, and exposed on both sides to 
the air current. It is therefore very sensitive as a 
thermometer ; cettainly much more so than the Bour- 
don tube or ordinary metal couple. 
I have always accepted the Continental records 
made in the winter as being free from any systematic 
error, but have iong felt that their summer ascents 
show temperatures that aic persistently too high. 
The policy of making all ascents at a fixed time, 
a.m. G.M.T., seems to me a most unfortunate one 
