222 
NEC ige 
[JuLy 6, 1899 
LETTERS TO THE EDITOR. 
[The Editor does not hold himself responsible for opinions ex- 
pressed 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 is taken of anonymous communications. | 
Magnetic Strain in Bismuth. 
Your report of the meeting of the Royal Society of Edin- 
burgh, held on May 16 (NATURE, June 22, p. 192), states that 
Dr. C. G. Knott has obtained a slight indication that there is 
a change of form in bismuth when strongly magnetised. 
In the Ph2l. Trans., vol. clxxix. (1888) A, p. 216, I de- 
scribed an experiment in which a rod of bismuth was found to 
exhibit an elongation of about 1°5 ten-millionths of its length in 
a magnetic field of 840 C,.G.S. units. As to the reality of this 
effect and the fact that it was due solely to magnetism, there 
was no doubt whatever. 
Since the publication of my paper I have repeated the experi- 
ment with another sample of bismuth obtained from Messrs. 
Johnson and Matthey; but though the field was brought up to 
nearly 1500 units, there was never the smallest indication of any 
magnetic change of length. An elongation one-tenth as great 
as that observed in the former case would have been easily 
perceptible. . 
After this experience I should hesitate to attach importance 
to any such observations unless the bismuth employed had been 
proved by analysis to be free from traces of magnetisable metals. 
SHELFORD BIDWELL. 
My experiments were made with a bismuth ze, the notion 
being that, as in like experiments with nickel tubes, any existing 
strain would be much more easily detected by means of secondary 
volume changes than by means of the direct elongation measure- 
ments which Mr. Bidwell so successfully carried out. Mr. 
Bidwell’s warning as to the necessity of having the material pure 
is well-timed. So far I have taken no special precautions in 
this direction ; but in the improved form in which I purpose re- 
peating the experiment, and from which I hope to get some 
really decisive result, this question of freedom from traces of 
strongly magnetic metals must of course be carefully considered. 
C. G. Knorr, 
Gooseberry Saw-fly. 
I sHALL be obliged if any reader of NATURE who has 
happened to pay attention to the gooseberry saw-fly will let me 
know whether my experience agrees with that of observers in 
other parts of the country. In Yorkshire the larve were so 
abundant in 1893, 1894 and 1895, that the bushes were in many 
places stripped of their leaves every summer. In 1896, there 
was a marked diminution, and many of the larve contained 
ichneumons. In 1897, 1898 and 1899, they have been so 
scarce that I have had difficulty in getting specimens for 
anatomical study. L. C. MIaqt, 
The Yorkshire College, Leeds, June 29. 
School Laboratory Plans. 
REFERRING to Mr. Richardson’s letter (p. 199), our labora- 
tory, now approaching completion, will afford, as regards 
chemistry in a room 30 by 26 feet, accommodation for twenty- 
seven boys, including one 18-foot bench for general purposes, 
and two draught cupboards. We have one 21-foot wall bench 
and two 18-foot double central benches in parallel, and one 
10-foot wall bench at right angles. I believe a novel feature to 
be the demonstrator’s platform placed on the top of and slung 
across the central benches, provided with a revolving chair and 
a table, and approached by steps. The whole sacrifices two 
working places only. The demonstrator has sixteen boys in 
front of him, five parallel with him, and six behind him, at a 
maximum distance of fifteen feet. is commanding position 
should save considerable time usually spent in running about. 
A large mirror might further aid matters. The central benches 
alone have reagent shelves, A. E. Munby. 
Felsted School, Essex. 
NO. 1549, VOL. 60] 
ILLUSTRATIONS OF MIMICRY AND COMMON 
WARNING COLOURS IN BUTTERFLIES. 
Na interesting, though brief, paper entitled ‘‘ Natura? 
Selection in the Lepidoptera” was read by Mr. 
Mark L. Sykes before the Manchester Microscopical 
Society on November 4, 1897, and published in the 
Transactions for the year (p. 54). The chief interest of 
the paper consists, as the author points out, in the eight 
excellent plates by which it is accompanied. These 
plates contain a very large number of figures reproduced 
by a photo-mechanical process from the insects them- 
selves. The author has evidently had at his disposal a 
very large and complete collection, and having selected a 
number of very fine illustrations he thus makes them 
available for all other naturalists. 
At the opening of his paper the writer expresses some 
doubt as to whether the subject is a suitable one for a 
Microscopical Society ; but on this question there need 
be no hesitation. The microscope is an instrument of 
the most varied uses, and is necessary in the investi- 
gation of this subject among others, for without its aid 
we cannot ascertain the depth to which mimetic resem- 
blance penetrates into the structure of organisms. The 
interpretation of these resemblances as due to natural 
selection implies that they are confined to visible effects,, 
and therefore the microscope should reveal an under- 
lying difference beneath the superficial similarity. Hence 
a paper which, by describing this fascinating subject, 
and abundantly illustrating it, directs attention to a 
promising field for microscopic inquiry, Is in every way 
suitable for the audience before which it was read, and 
the Manchester Society is to be congratulated upon the 
broad view it has taken of its subject and responsibilities. 
The present writer has already commenced the study of 
mimetic resemblance from this point of view, and has. 
found that the methods by which the transparency. 
which is necessary for the likeness attained in a group of 
Lepidoptera from South America differ in the most 
marked degree, although the superficial resemblance is. 
of a very high order (/ourn. Linn. Soc., vol. xxvi., 1898, 
pp. 596-602 ; plates 42, 43, 44). 
The great interest of Mr. Sykes’ paper is the abundant 
illustration which it provides for the two different classes. 
of resemblance often confused together under the name 
of “Mimicry.” A few words on the history of the 
recognition and suggested explanation of these two 
classes may be useful, inasmuch as great confusion still 
exists upon the subject. The theory of mimicry was. 
first suggested by H. W. Bates in his important paper 
published in the Zyansactzons of the Linnean Society 
for 1862 (vol. xxili.): He here suggested the idea of a 
conspicuous, abundant, and specially defended species, 
serving as the model towards which other com- 
paratively rare and defenceless species are brought by 
natural selection. His illustrations were taken from the 
fauna of tropical America, and the explanation was sug- 
gested to him by the study of his collection after his 
return home from his prolonged visit to the Amazon 
Valley. The theory is of special interest, as it was prob- 
ably the first great result of the theory of naturab 
selection after its appearance in the Journal of the 
Linnean Society in 1858, and in the “ Origin of Species” 
in 1859. Bates’ generalisation was extended by A. R. 
Wallace to the tropical East (7vass. Linn. Soc., 1866, 
vol. xxv.), and by Roland Trimen to Africa (Zyazzs. 
Linn. Soc., 1870, vol. xxxi.). In the first-named paper 
Bates expressly pointed out that his explanation did not 
cover all cases of mimetic resemblance, but that there 
were a very large number of species abundant in indi- 
viduals, and presumably specially defended, which never- 
theless “mimic” each other. Furthermore, this kind of 
resemblance is as close and detailed as that which the 
Batesian theory of mimicry sought to explain. For such 
