
Marcu 17, 1923] 
NATURE 
359 

by no means easy or altogether certain, and hence the 
microscopic method should have been invaluable. 
‘The feaulte of these experiments were obtained too 
late to be used on a large scale, but they certainly 
ow an example of the class of information 
obtainable. 
_ Another application of a rather humorous nature 
can also be cited. A large consignment arrived in 
the factory of what was called ground silica. This 
was sent to the analytical department and its per- 
centage of silica ascertained. It was pointed out, 
however, that the value of ground silica did not lie in 
the purity of the material but in its fineness and 
homogeneity. This was tested both by elutriation 
and the microscope. The microscope revealed the 
fact that the material was nothing but an inferior 
sand, with very little grinding. Ground silica must, 
of course, always show conchoidal fracture and not 
rounded grains. This sample was also shown to 
contain the mineral glauconite actually replacing the 
tests in small fragments of foraminifera, and hence 
had during its formation been closely connected with 
thesea. The price per ton indicated that the material 
was supposed to have been obtained by the grinding 
of vein quartz. 
The identification of asbestos has been published 
before, but it bears repeating. Platinised asbestos 
was extensively used in the sulphuric acid plant. 
The asbestos was originally supplied from the conti- 
nent, but during the war this supply was not available. 
The South African asbestos or the mineral crocidolite 
was used as a substitute with very disastrous results. 
It was decided, therefore, that the nature of the 
original asbestos must be obtained by chemical 
analysis and all samples similarly tested. The 
chemical analysis of a complicated silicate like 
asbestos is a long and by no means easy process, as 
_the asbestos is seldom free from other complicated 
silicates. Now, it was found by a very simple 
mineralogical test that the original sample was the 
mineral chrysotile, and by similar tests it was quite 
easy to ascertain which of the other samples was also 
chrysotile and to pick out the purest. In this way, 
a dozen samples were tested in two hours, whereas 
the chemical analysis had already been in hand for 
three months, and was likely at the same rate to take 
another six and give no information whatever. 
The simple test for chrysotile was mounting it on 
a microscopic slide in mononitrobenzene and rotating 
it between crossed nicols. The refractive index was 
obtained by the Becke method. The refractive index 
together with the birefringence and optical character 
render the mineral quite distinct from any other sold 
as asbestos. These are three of the very many 
occasions that cropped up so frequently. 
ASHLEY G. LOWNDEs. 
Marlborough College, Wilts. 

Factors of Odorous Strength. 
In the letter from Mr. J. H. Kenneth published in 
Nature of February 3, page 151, a relation is in- 
dicated between the odours of certain substances and 
specific gravity. If, however, we examine the boiling- 
points of the odorous constituents of the four oils 
mentioned, we find that in order of increasing vapour 
pressure the oils stand as follows : sandalwood (305), 
cedarwood (280), origanum (230), and terebene (160), 
the figures in brackets being the approximate boiling 
points. This order is precisely that represented by 
the specific gravity quoted by Mr. Kenneth. 
I scarcely think the phenomenon with which 
Mr. Kenneth’s letter deals, can safely be ascribed to 
NO. 2785, VOL. IIT] 
a et 

the specific gravity of the oils, although possibly in 
this instance the specific gravity is a property con- 
comitant with volatility. Volatility alone, however, 
does not afford a completely satisfactory explanation 
of this and many other phenomena connected with 
the smell of an odoriferous substance. There are at 
least four factors concerned, namely: (1) Volatility ; 
(2) solubility in the aqueous layers in the nose ; 
(3) solubility in the lipoid fats of the nose, and 
(4) chemical reaction with osmoceptors in the nose. 
A substance which fails to satisfy any one or more 
of these factors is odourless, and it is obvious that 
variations in the factors will produce variations in 
both the strength and the quality of the odour. 
T. H. Durrans. 
The Dyson Perrins Laboratory, 
South Parks Road, 
Oxford, February 6. 

With reference to Mr. Kenneth’s letter in NATURE 
of February 3, p. 151, I should like to point out 
that, if the ‘‘ votes’’ be counted on a sort of pro- 
portional representation scheme by adding to the 
first votes for each substance half the second, a 
third of the third, and a quarter of the fourth, we 
get the following results in votes : 
S. C. O. Bs 
18-66 11°56 9°33 6°33 
This result seems to me to enforce Mr. Kenneth’s 
argument. FRANK H. PERRYCOSTE. 
Higher Shute Cottage, Polperro R.S.O., 
Cornwall, February 17. 
The Life-Cycle of the Eel in Relation to Wegener’s 
Hypothesis. 
In NaturRE of January 27, p. 131, under the title 
‘The Distribution of Life in the Southern Hemi- 
sphere, and its Bearing on Wegener’s Hypothesis,” 
an account is given of a discussion at a recent meeting 
of the Royal Society of South Africa on this question. 
Opinions were divided, the geologists suspending 
judgment, while the hypothesis was opposed on 
botanical and entomological grounds as being un- 
necessary. On the other hand, it was said that the 
most important zoological evidence in support of 
Wegener’s theory was provided by the distribution 
of the isopod, Phreatoicus. 
It seems to me that strong evidence in favour of 
Wegener’s hypothesis is to be found in the life- 
history of the European freshwater eel, as revealed 
by the brilliant researches of Dr. J. Schmidt, of 
Copenhagen. For something like eighteen years 
Dr. Schmidt has been engaged on this subject. He 
has published numerous papers and has summarised 
his results in the Philosophical Transactions, pub- 
lished a year ago, and quite lately in NaTuRE 
(January 13). It will be sufficient for the present 
purpose to allude only to certain of his results. 
Of the two freshwater eels of the North Atlantic, 
the American species spawns somewhat to the south 
and west of the spawning region of the European 
species, and the larve attain full size and, after 
metamorphosis, enter the freshwaters of the American 
coast when about one year old. On the other hand, 
the larve of the European species, originating more 
to the east but still in the same region, are trans- 
ported by the Atlantic Drift and its continuations, 
aided perhaps by their own efforts, it may be for 
thousands of miles, as shown in the chart, p. 51 of 
the article in Nature of January 13. Still more to 
the point, the larve are about three years old when 
they become transformed into elvers and enter 

