316 
edition, is incomprehensible, saving the assumption that 
both with respect to his father’s annual reports 
and other sources, the son was wholly incapable of 
doing his father justice. It is a pity that the 
task of preparing a second German edition was 
not entrusted to a competent botanist, because the 
original work, apart from the uncompromising antagonism 
to Evolution that pervades it, still occupies an undisputed 
position in modern botanical literature. As it is, the 
French edition is not merely an advance on the original 
German—it is incomparably better than the second 
German edition. Itis only, however, fair that some justi- 
fication of such assertions should be given. Taking the 
chapter on Oceanic Islands as an example, it may be 
confidently stated that no additional information is given ; 
yet there is no branch of geographical botany that has 
advanced more during the last decade than insular. On 
the other hand Tchihatcheff embodies nearly all that was 
known up to date. One slight alteration observed in this 
chapter is—Madeira is stated to be 50 German geographi- 
cal miles nearer Europe than the Azores, instead of 150, 
as inthe original. Then certain unfounded statements in 
refutation of the arguments of other botanists concerning 
the relationships of insular floras remain uncorrected. 
Thus, in allusion to Sir Joseph Hooker’s demonstration 
(“Insular Floras,” p. 7) that the vegetation of St. Helena 
has, on the whole, its nearest affinities in South Africa, it 
is objected, on the authority of Roxburgh, that three out 
of the five genera named by Hooker were originally in- 
troduced into the island from the Cape of Good Hope, 
whereas an examination of Roxburgh’s enumeration of 
the plants of St. Helena reveals the fact that the indi- 
genous, and endemic, St. Helena species of the genera in 
question were unknown to him, and his remarks apply 
only to actually introduced species. Again, to repeat 
in 1884 such statements as that the vegetation of Juan 
Fernandez has little systematic relationship with that of 
the Chilian or Antarctic floras and that Pringlea anti- 
scorbutica is restricted to Kerguelen Island is unpardon- 
able, because the contrary is now historical. Defects 
such as those pointed out are numerous, but as they 
are mostly due to the state of knowledge fifteen years ago, 
the author of the work of that date is not to be blamed 
for them; rather the present editor and publisher for 
offering the public an old book as new. 
W. BOTTING HEMSLEY 
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. 
No notice ts taken of anonymous communications. 
[The Editor urgently requests correspondents to keep their letters 
as short as possible, The pressure on his space is so great 
that it ts impossible otherwise to insure the appearance even 
of communications containing tnteresting and novel facts. | 
Nomenclature in Elasticity 
THE word sfress is used, sometimes in the sense of oad, some- 
times in that of /oad per unit area. Clearness, however, 
requires these two ideas to be kept perfectly distinct, and there- 
fore to be denoted by separate terms. oad is surely expressive 
enough, or, if not, there is the more comprehensive word force : 
why then use séress synonymously? It would be far better to 
reserve stress to signify load per unit area. This Prof. Kennedy 
(p. 269) calls xfensty of stress ; but why not sévess simply? The 
NATURE 
[August 6, 1885 
word zéenszty is not in itself suggestive of anything distinctive, 
and is therefore useless. 
Pressure and tension are terms used in the same loose manner, 
though, when intended to represent force, they sometimes have 
the word who/e prefixed. Is it not better to say force when we 
mean force ? We can then reserve pressure and /ension as vector- 
synonyms of s¢vess in the sense of force per unit area, which is 
indeed their usual 7éVe. 
Another misused term is vesz/ience, which sometimes denotes 
the worvé done in producing froof strain in a body (Rankine’s 
definition), sometimes the work done fer unit volume in produc- 
ing roof strain, sometimes the work done fer unit volume in 
producing azy strain. I prefer, myself, the third definition : the 
second would then be the proof vest/ience, and the first might be 
called the strain-energy. 
However, whatever terminology is finally agreed upon, let it 
be perfectly definite and consistent. 
In his Fig. 1 (p. 269) Prof. Kennedy writes: ‘* Breaking 
load, 18°85 tons per square inch.” According to his own 
nomenclature, he should surely say: ‘‘zntensity of breaking 
stress 18°85 tons per square inch,” and this I should prefer to 
call simply the breaking stress—premising that for ¢ozs I should 
write fons’ weight. In this case, as the diameter is } inch, and 
therefore the section "442 square inch, the breaking /oad is 8°33 
tons’ weight. Similarly in the other figures. 
Christ Church, Oxford Rogpert E. BAYNES 
Earthquake-Proof Buildings 
Mr. Muir is quite correct as to the facts and date of the 
introduction cf the aseismatic tables into Japan. In 1869-70 
seven aseismatic tables for carrying the lighting apparatus were 
sent from here and erected in Japan, and Mr. Simpkins, who 
has recently returned from Japan, informs me that there are 
three in action at present. Two iron towers, 46 feet high, with 
this arrangement at their base, were also constructed and shipped 
for Japan, but the vessel was lost and no more were sent out, as 
the engineer in charge—Mr. Brunton—took an unfavourable 
view of their efficiency—his idea being that they would not 
work, as he considered that buildings of ‘‘ great weight and 
solidity, thereby adding to their inertia and checking their oscil- 
lation, were best suited to meet the difficulty in Japan.” Mr 
Milne’s experiments with aseismatic tables have borne out Mr. 
David Stevenson’s original view as to their power of mitigating an 
earthquake shock. For fuller information see NATURE, vol. 
Xxx. Pp. 193. D. A. STEVENSON 
Edinburgh, August 3 
A Mechanical Telephone 
HAVING observed in this week’s NATURE a notice of a 
“‘mechanical telephone” said to be brought from America, I 
may state that so far back as 1878 I experimented on the trans- 
mission of sounds by wires, and communicated the results ob- 
tained, from a large number of experiments, to the Physical 
Society of London in March, 1878 ; the paper being afterwards 
published in the P/zlosophical Magazine for August, 1878. 
These experiments are referred to by the Count du Moncel in 
his book on ‘* The Telephone,” published in 1879. I found no 
difficulty in carrying on a conversation through wires laid in 
various ways from room to room of a house; and musical 
sounds, breathing, and whistling were also readily transmitted, 
and through most unlikely arrangements, such as a common wire 
fence. Various materials were tried for the transmitting and re- 
ceiving ends—disks of cardboard set in deepish rims being found 
to give excellent results with a No. 16 copper wire. In one of 
my experiments I found that the disks were not required, the 
wire itself picking up and transmitting the sounds. ‘The results 
obtained were most interesting ; but as the range was necessarily 
limited, it did not seem to me that there was much scope for 
practical application. W. J. MILLAR 
100, Wellington Street, Glasgow, July 31 
Electrical Phenomenon 
AxourT ten o'clock in the evening of July 23 a party of four of 
us were standing at the head of the avenue leading to this house, 
when we saw a feebly-luminous flash appear on the ground at a 
distance of some thirty yards down the avenue. It rushed 
towards us with a wave-like motion, at a rate which I estimate 
at thirty miles an hour, and seemed to enyelop us for an instant. 
