Marcu 11, 1897] 
NATURE 
439 
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 ts taken of anonymous communications. | | 
Dynamical Units. 
WirHIn the moderate dimensions of a letter it is hard to | 
give due weight to every aspect of a complicated matter, and 
while trying to emphasise one side I have somewhat overstated 
the case, as is evident from the way in which Prof. Lodge has 
taken me up. I was only considering the teaching of elemen- 
tary dynamics to engineering students. Ido not object to a 
teacher explaining that inertia is such an important constant 
property of matter, that equality of inertia is our definetion of 
equal quantities of matter. What I do object to is, a common 
inversion of this, by which equality of inertia is exp/ained by 
saying that the quantities of matter are equal. In addition, I | 
urge that teachers of elementary dyzamzcs should call what is 
usually called mass, inertia, so as constantly to bring before the 
student the fact that this is ¢4e property with which the 
dynamics of motion deals. I do not plead guilty in this to con- 
fusing the issues. The issues of Prof. Perry’s review have been 
overlaid with a discussion as to one of the greatest advances of 
modern physics, namely the possibility of representing physical | 
quantities by algebraic symbols ; but I was trying to recall the | 
original issue, as to the way dynamics should be taught to 
engineering students. Babes must be treated babyishly, and as 
long as engineering students are what they are now, and have 
to attend a variety of lecturers, and read engineering books as 
they are, I agree with Prof. Perry in recommending that the 
engineer's unit of inertia be used by their teachers. I have 
already explained that a multiplicity of units isa very minor 
difficulty to those who have once grasped what it is that is being | 
measured, but I do think it confuses them, while getting these 
ideas, for one teacher to use one system, and another another 
system, and for each teacher to call the system of the others by 
hard names. Gro. FRAS. FITZGERALD. 
Definite Variations. 
Mr. F. A. BATHER, in the January number of Matural 
Science, has some remarks on Prof. Cope’s ‘* Factors of Evolu- 
tion ” which seem to call for further comment. The case is cited 
of sheep taken from Ohio to Texas losing the fine quality of their 
wool, and this definite variation, due to environment, being 
apparently inherited and cumulative, in spite of selection by the 
breeders of those lambs which least present the new character. 
Such facts as these are not new, and it seems to me that they 
represent simply a phase of atavism. On July 23, 1890, I was 
present at a meeting of the Royal Horticultural Society, at 
Chiswick, and heard Mr. E. J. Lowe give’an interesting address 
on ferns. In the course of it, he told how he had a great num- 
ber of varieties of the hart’s-tongue fern, which, on changing his 
place of residence, he moved into new and poorer soil, They | 
all reverted to the typical form, and it was not until they were 
again transplanted to good soil that they consented to exhibit 
their varietal characters! Now in the case of the sheep, the 
fine wool of the Ohio breed is not a specific character, but a 
varietal one produced under domestication ; and it is not sur- 
prising, therefore, that removal to a locality less favourable, and, 
-perhaps, more resembling that of the original type of the species, 
should produce reversion. But it is probable that, as in the 
case of Mr. Lowe’s ferns, the varietal character could be made 
to reappear by transference to the former kind of environment. 
The precise explanation of such facts as these may probably be 
found in Dr. Weismann’s principle of germinal selection, which 
has surely been more or less understood for a long time. The 
sheep is born with two or three distinct possibilities, as to its 
wool; one locality favours one of these possible developments, 
one another. It is a case parallel to that of an amphibious 
“Ranunculus, which can be made to assume one form or the 
other, according to the terrestrial or aquatic environment. 
It is worth while to add, that here in New Mexico, one fre- 
quently sees small, usually pale yellowish-brown, horses, with 
extremely well-marked leg-stripes. These are descendants of 
the horses which ran wild in former years over this country ; 
and there can be little doubt, I think, that they represent an 
atavistic variety. 
NO. 1428, VOL. 55 | 
| cells. 
While it is probable that really new variations are equally in 
all directions, practically the variation of most organisms is 
remarkably definite, because so largely atavistic. And these 
definite atavistic variations may be perpetuated, in new combina- 
flons, in new races. It is precisely this which gives rise to 
“kaleidoscopic characters” in a group. A character may ap- 
| pear here and there, and species may be represented by different 
combinations of the same characters, as words are composed of 
combinations of the same letters. 
To cite an illustrative instance, the wings of bees present 
frequently one marginal, three submarginal, and three discoidal 
The submarginals may be reduced to two, or even to 
one, and the discoidals to two. All sorts of combinations, as to 
the number, shape, and size of these cells will be found, but the 
marginal will 7zo¢ be found lacking, nor the first discoidal or first 
submarginal absent, nor will the number of submarginals be 
found increased.’ Really new variations, as new ones running 
out to form a second marginal or a fourth submarginal, appear 
in slight degree, but doubtless sufficiently to afford material for 
selection, under new environment; but the old and common 
variations may occur suddenly, so that there may even be a 
radical difference between the opposite wings of the same speci- 
men. Very rarely, a remarkable sport will occur, not in accord- 
ance with our expectations, but these are much too rare to have 
selective value. In other hymenoptera, as the sawflies, the 
range of common variation is quite different; but still the 
variations to be looked for in each family are not miscellaneous, 
but run along certain well-recognised lines. To show that new, 
not atavistic, variations take definite lines is another matter, 
and I do not believe it can be done. 
T. D. A. COCKERELL. 
Mesilla, New Mexico, U.S.A., February 18. 
The Coral Reef at Funafuti. 
THE report on the coral reef at Funafuti that was read to the 
Royal Society on February 11, will doubtless be of very great 
interest to all who have studied the very difficult problems con- 
cerning the origin of reefs and atolls. 
At the same time, many will wonder why Prof. Sollas charac- 
terises the boring as a failure. Scientific expeditions very rarely 
accomplish all that is anticipated or even expected of them, but 
they are not necessarily failures in consequence. It is true that 
the borings at Funafuti could not be carried to a depth of more 
than 105 feet, and that the structure they revealed was not ‘* what 
a field geologist might have anticipated”; but they revealed 
the very important fact that underlying a coral reef of 50 feet in 
thickness, there was a stratum of sand containing a few coral 
blocks. 
It is perhaps premature to consider, until further details are 
published, whether this fact supports the views of Mr. Murray or 
his followers ; but what is perfectly clear at once, is that it lends 
no support to the well-known subsidence theory. 
I think it is of importance to call the attention of the scientific 
public to this at once, because, after dismissing the boring as a 
failure, Prof. Sollas gives the results of the soundings made in the 
neighbourhood of the island by H.M.S. Penguzz, and concludes 
by the statement that, in his opinion, these soundings support 
Darwin’s theory of coral atolls. 
I should not like at this stage to take upon myself the respon- 
sibility of saying they do not, but I should like to ask, after the 
negative evidence afforded by the borings, upon what grounds 
Prof. Sollas bases his opinions, SypNeEY J. HIcKson. 
Owens College, Manchester, February 20. 
Two Unfelt Earthquakes. 
REFERRING to Prof. John Milne’s interesting communication 
on “Two Unfelt Earthquakes,”’ asking for information as to 
whether these disturbances have been instrumentally recorded 
| elsewhere, Dr. Copeland requests me to say that an examination 
of the photographs of the oscillation-curve of the bifilar pen- 
dulum at this observatory shows several disturbances on February 
7, the first of the dates mentioned by Prof. Milne. These are 
as follows:—At 7.37 a.m. an abrupt movement of the pen- 
dulum towards the north; from 8.24 to 8.41 a.m. a distinct 
reduction in the intensity of the colour of the photographic 
1 Melipona and Trigona are exceptional, and in many ways depart widely 
from the normal type, so that they hardly come within the range of typical 
bee-modification. 
