Aprit 18; 1884.] 
for. In my third and last’ notice (iii. 143) the man- 
ner in which the muscles attached to the occipital 
style are inserted was alluded to, and it was compared 
with an ossified ligamentum nuchal. All of this 
T still maintain. . At that time, for lack of material, I 
had not especially looked into its physiology; and my 
-discussion with Mr. Jeffries closed (Feb. 8, 1884). 
Since, both through my reading and observation, 
much has come to my notice of interest with regard 
toit. Garrod’s dissections of Plotus anhinga are very 
suggestive. Dr. Gill had kindly called my attention 
to Yarrell’s paper, before his notice in Science ap- 
peared, which he had unexpectedly come across 
‘while searching for facts to illustrate another subject. 
‘Finally, in one of the most useful and reliable of 
books, Coues’ ‘Ornithological bibliography,’ I had 
noticed Rudolphi’s article; but other matters were en- 
gaging my attention then, and reference was not made 
to it. Therearestill others. I have already cited Ky- 
ton’s figure (iii. 143), and believe, at the time Dr. 
Gill’s review of my work appeared, I was hardly en- 
titled to the charge he brings against me init. lam 
more and more convinced, every day of my life, that 
good illustrations of such common facts in anatomy 
are most urgently demanded. R. W. SHUFELDT. 
A singular optical phenomenon. 
I think it would well repay almost any one to study 
the beautiful phenomenon so clearly described by 
“F.J. 8.’ (Science, No. 57, p. 275), and so suggestively 
discussed by Professor LeConte (No. 61, p. 404+). My 
own theory of it involves no inverting action, as 
in the camera, and no primary dependence upon 
binocular vision, but, rather, it resembles the theory 
of watered silks, or of chords and beats in music. 
It seems to me geometrically demonstrable; and it 
‘includes the phantom meshes’ gigantic size, their 
bewildering motions, their conspicuousness even to 
eyes out of focus for the actual wires, and the non- 
appearance in them of objects attached to those 
wires. 
Before the observer are two parallel screens of 
Sjuare-meshed wire netting. The-coarser one is seen 
through the finer, and the two are at distances from 
him nearly proportional to the diameters of their 
meshes, measured from centre to centre of the wires. 
To fix the ideas, suppose that he looks with only one 
eye, seeing the nearer wires black and the farther 
ones bright: then, if the above proportionality be 
exact, all the bright wires can be simultaneously 
eclipsed, each by a separate dark wire; or, upon mov- 
ing the eye very slightly to the right and upward, all 
the bright wires will flash into view at once. Now 
let the observer advance or retire a few inches from 
this first position, so that the dark wires may subtend 
visual angles a little larger or smaller than do the 
corresponding bright ones: several successive bright 
wires will thus be in view, then one or more will be 
eclipsed, then several others will be seen, and so on; 
that is, the phantom screen will be formed, with its 
great square meshes and shadowy bars. 
Next let the observer move slightly to the right: 
the phantom also moves, but more, and to the right 
or the left, according as he is in front of or behind his 
first position. Indeed, the motions of the phantom 
bars, and the visual angles they subtend, areas if the 
observer viewed a virtual image whose plane passed 
through his first position, but imagined it to be some 
feet in front of him. The size of the virtual image 
would be very nearly such, that, in it and the farther 
screen together, there would be as many bars to the 
foot as in the nearer screen. Its colors would appear 
SCIENCE. 
475 
to be those of the farther screen, but weaker and 
oppositely arranged. It would not be upside down. 
Indeed, if ‘F. J. S.’ will paint the upper wires of the 
farther screen vermilion, or will hang behind them 
a blue curtain, then I think that the upper meshes, 
but not the bars, of the phantom, will be reddened; 
or the upper bars, and more slightly the meshes, of 
the phantom, will be bluish. Or, if he will paint the 
vertical wires red and the horizontal wires yellow, 
probably the phantom meshes will incline to orange, 
the vertical phantom bars to yellow, and the horizon- 
tal ones to red. 
Suppose that two-thirds of the light coming from 
within the boundary of the farther screen be from 
the bright wires: then the phantom meshes will be 
three times as bright as the phantom bars; but at 
their edges they may blend into one another, the 
eclipses there being less complete. Thus no lines 
appear in the phantom whose pictures on the retina 
are not much broader than the picture of a point, even 
when out of focus, and hence the phantom is seen 
by near-sighted and far-sighted alike. 
Phantoms often less simple and conspicuous may 
be got when the visual angles subtended by single 
spaces in the two screens are not approximately equal, 
but are approximately in a simple numerical ratio. 
The screens may also be of lattice-work, or pale fences, 
not necessarily parallel, seen two or three deep 
against the sky; and the effects are sometimes very 
beautiful. 
Undoubtedly, when the screens are fine, binocular 
vision, with the stereoscopic matching of patterns, 
comes in, as suggested by Professor LeConte; making 
the phantom seem real and solid, and fixing its as- 
sumed distance from the observer. But I leave this 
part of the discussion to him, because he can treat it 
far better than I can. JAMES EDWARD OLIVER. 
Cornell university, April 8. 
I was gratified to find that the phenomenon de- 
scribed in No. 57 proved of interest to Professor 
Joseph LeConte. He states that my explanation of 
the cause of the phenomenon is erroneous, and I am 
in no wise qualified to dispute him. Nevertheless, a 
careful repetition of the experiment would indicate 
that his explanation is not the correct one. The 
phantom image is as readily seen with one eye as 
with two; and I still think I am correct in saying it 
is inverted and magnified. I hope Professor LeConte 
will make the experiment himself, and give us his 
explanation of the phenomenon. In the mean time, 
allow me to state the facts as they occurred in an 
experiment made after reading his letter. 
Standing about twelve feet from an ordinary fly- 
screen, and looking through it at the blinds of a house 
about one hundred and fifty feet distant, phantom 
lines, alternately a light one and a dark one, are seen 
crossing so much of the field of view in which the 
blinds he, but not continued beyond their limits. 
The lines remain visible, although one eye be closed. 
The image rises as I bow my head, and sinks as I 
lift it. Is not this evidence of inversion? 
I can readily count the lines that lie across a blind, 
twelve light and twelve dark ones; but, in order to 
correctly count the actual slats in the blind, I am 
obliged, on account of the distance, to have recourse 
to a telescope. My wife, who is short-sighted, can 
only distinguish the mere outline of the actual blind; 
but the phantom lines are plainly visible to her. The 
number of slats in a blind is thirty, which would give 
sixty alternating dark and light lines. Is not this 
evidence of magnification? F. J.S. 
