I? hsKUARY 21, 1907 | 
NWALORE 
387 
graphically many of the properties of a train of 
lenses, especially points of magnification and aperture, 
but it must be used discreetly. The author has him- 
self been betrayed into an erroneous conclusion by 
means of it. If the origin of such a catena is a point 
of light, and a small opaque disc is inserted in the 
opening limb, the lane of darkness succeeding it will 
follow the same rules of formation as the cone of 
light, i.e. it will be a succession of cones having 
bases at the apertures and vertices at the images. 
One base is the conical projection merely of the pre- 
vious one. Now Sir A. E. Wright is rather prone 
to overrate the resemblance between projections and 
images. He calls the similitude of a candle shining 
through a small hole upon a screen an image. That 
is not the sense in which the word is technically 
employed. An image is always a focussed image, 
though the qualification is not invariably stated. But 
the confusion of the two ideas leads the author into 
language which cannot be interpreted otherwise, we 
think, than as implying that every detail in one aper- 
ture, say a fleck of dust, is repeated as a genuinely 
focussed image situated at succeeding apertures. It 
is only a projection that takes place in such a case, 
and it would occur equally well at any plane along 
the line; and when the source of light has finite 
dimensions, even this strictly defined projection will 
not occur. This idea that everything in one aperture 
is focussed in succeeding apertures leads the author 
to more than one conclusion which, if we read him 
aright, is not sound. 
An example of error arising really from the neglect 
of the focussing idea occurs in the author’s justifica- 
tion, for it is not a proof, of the expression for the 
numerical aperture, in so far as it increases with the 
index of refraction of the external medium. In one 
of the diagrams (p. 74) the final surface of a convex 
lens is drawn as a plane, and in that case the excentric 
ray of a convergent beam will not intersect the axis, if 
it finds itself in water or in oil, at a point so close 
to the lens as if air were the medium in which it 
emerges. That is true, but it is no proof or even 
illustration of the point under consideration, which is 
the effect of the index upon the numerical aperture. 
This will be manifest by supposing the lens a convex 
meniscus with the second surface concave, and the 
origin of light to be at the point which is conjugate 
to the centre of curvature of the second surface. 
Then the light, both immediately before and after 
encountering the final face, will be normal to it, and 
an emergent ray will intersect the axis at the same 
point whether the surrounding medium be air or 
water. But the numerical aperture would still be 
affected by the medium. 
The aperture question seems, indeed, a stumbling- 
block to the author. He knows as well as others 
do the connection between the radius of the false disc 
and the numerical aperture, and he rightly defines the 
latter, but his proof of the proposition at pp. rio, 111, 
would make the numerical aperture proportional to 
the tangent (instead of the sine) of the semi-angle of 
the cone of light. : 
NO. 1947, VOL. 75] 
The author has constructed an instrument to which 
he has given the name eikonometer, not a very happy 
one, for its object is, not to measure images generally, 
but by measuring certain images to arrive at magnify- 
ing. powers. The principle, which is not so new as 
the author supposes, any more than the constructed 
instrument, is the fact that if two lenses are placed 
upon the same axis at any convenient distance apart, 
the first principal focus of one is conjugate to the 
second principal focus of the other, and the object 
bears to the image the same linear proportion which 
the focal lengths of the two lenses bear one to the 
other. Hence, if the object be of known dimensions, 
and its image be measured and therefore also known, 
and if one of the focal lengths be known, the other 
is also at once determinable. The actual focal length 
of an entire microscope may thus be, and has been. 
determined in one observation. The magnifying 
power is usually taken as the number resulting from 
the division of 250 by the focal length in millimetres, 
but this is an arbitrary rule which presupposes that 
250 millimetres is the least distance of distinct vision, 
which is certainly not the case universally. The 
author does not use the eikonometer quite logically. 
In finding the focal length of a microscope his in- 
struction is, first to focus the microscope in the usual 
way upon a scale of known dimensions situated on 
the stage, then to place the eikonometer over the eve 
end of the instrument and to read off. 
The instruction should be first to place the eilkon- 
ometer in position, and then by means of the 
ordinary focussing arrangement of the microscope 
bring the image of the scale on the stage into 
position at the scale of the eikonometer, and then 
read off. 
The author thinks himself at issue with Abbe in the 
conclusion drawn from the grating experiment of the 
latter. This arises through a misapprehension, and 
the experiments which the author cites do not affect 
the conclusions which Abbe reached. 
Tuomas H. BLakestey. 
THE CRUSTACEA OF DEVON 
CORNWALL. 
The Crustacea of Devon and Cornwall. By Canon 
A. M. Norman, F.R.S., and Dr. Thomas Scott. 
Pp. xv+232; 24 plates. (London: William Wesley 
and Son, 1906.) Price 1i/. 4s. net. 
TUDENTS of British carcinology owe their 
thanks to Dr. Norman and Dr. Scott for the 
first appearance of a volume embracing the wide field 
of our indigenous crustacean fauna, as hitherto re- 
corded in the Devon and Cornwall area. Dr. 
Norman’s researches in this district, extending over 
a period of more than fifty years, are too well known 
to need mention here, and the publication of his 
records is a welcome addition to our literature. The 
introduction contains some interesting comparative 
tables on the distribution of species north and south 
relative to the area with which the work 
cerned. The body of the work comprises an enumer- 
AND 
is con- 
