Marcu 7, 1907 | 
INA IM OG IE, 
439 
peroxide, inhibits rusting. He ignores the fact that there 
are other substances, such as potassium iodide, which 
immediately destroy hydrogen peroxide and yet do not 
inhibit the rusting of iron. Moreover, if Prof. Dunstan’s 
assumption that substances which destroy hydrogen 
peroxide (which he regards as an essential initial product 
of rusting) inhibit rusting be accepted, it will be necessary 
to admit, contrary to the general experience of chemists, 
that the presence of a substance capable of removing one 
of the products of an action does not accelerate the action, 
but actually prevents it. 
Prof. Dunstan does not say in what respects his experi- 
ments on the oxidation of iron have afforded results differ- 
ing from my own, but I may remind him that only after 
repeated failures was I successful in bringing together iron, 
oxygen, and water, and in avoiding the presence of acid. 
: GreraLtp T. Moopy. 
Central Technical College, February 22. 
The Valparaiso Earthquake, August 17, 1906. 
Pror. Mitnr’s note in Nature of February 21 raises 
an interesting question which can readily be answered ; 
the earthquake which preceded the Valparaiso shock 
originated under the North Pacific Ocean in about 30° N. 
lat., 170° E. long., at about oh. 11m. a.m. G.M.T., or 
352 minutes before the Chilian earthquake as recorded at 
Santiago. This position does not agree with the distance 
given in the note, but Prof. Milne, in correspondence, has 
informed me that this is in error, and the distance, as 
indicated by the Shide diagram, is 90°, which is in close 
accordance with my own determination of the distance. 
It must be remembered that all attempts at deducing 
the distance of origin from a single seismogram are neces- 
sarily approximate, though the error will probably be 
within 5° of arc, or about 350 miles, in the case of a 
great earthquake giving a complete record. The determin- 
ation of the place of origin becomes easy when a sufficient 
number of records from widely separated localities are 
available, and these are at my disposal, for, seeing that 
the Chilian earthquake was likely to be an important one 
in connection with an investigation on which I was 
engaged, I wrote to a number of seismological stations 
the addresses of which were known to me, and met with 
a most generous response to my requests. Unfortunately, 
when the copies of seismograms came in it was evident 
that they recorded two earthquakes, of which the earlier 
was of unknown origin, the record of which in every case 
overlapped that of the Chilian one, and rendered the latter 
practically useless. R. D. Otpuam. 
Nomenclature of the Proteins. 
In the current number of the Proceedings of the 
Chemical Society, the council has issued some valuable 
proposals for change in the nomenclature of the proteids 
and allied substances. While not venturing to criticise the 
majority of the recommendations, I notice a definition in 
the proposed subclass 5 which appears to me slightly in- 
accurate. The subclass in question reads as follows :— 
““s. Sclero-proteins. This new word takes the place of 
the word albuminoid in the limited sense in which the 
majority of physiologists have been accustomed to use it. 
It includes such substances as gelatin and keratin; the 
prefix indicates the skeletal origin and often insoluble 
nature of its members.” 
Now, it seems to be a generally accepted view that 
gelatin does not exist ready-formed in nature, but results 
from the hydrolysis or hydration of collagens (v. Allen’s 
“Organic Analysis,’’ vol. iv., and Cohnheim’s ‘‘ Chemie 
der Eiweisskérper ’’). Is not gelatin as much a_ product 
of protein hydrolysis as acid-albumin or allkali-albumin, 
for which the generic term meta-proteins is now proposed ? 
Would it not, therefore, be preferable to reserve the term 
sclero-proteins, in its strictest sense, for the wholly in- 
soluble products of animal-cell activity, such as chondrigen, 
ossein, sericin, and keratin, and class their hydration- 
products such as gelatin and silk-gelatin among the meta- 
proteins? 
' The committee apparently sees no objection to including 
gelatose among the proteoses. W. S. GILLEs. 
Bocking, Braintree, Essex, March 4. 
NO. 1949. VOL. 75] 
Maximum Gravitational Attraction on a Solid. 
Can you tell or refer me to the solution of the following 
question :— 
What will be the shape of a definite quantity of mass 
of given specific gravity in order to obtain maximum 
gravitational attraction at a point on its surface? I have 
tried various shapes of equal volume, including square 
and rectangular figures, hemisphere, sphere, and cones. 
For these shapes I found that the maximum attraction 
obtained at the centre of the base of a cone the apex 
angle of which was about forty degrees; no doubt the 
frustrum of such a cone would attract with greater force. 
This question is no doubt of academical interest only, 
but the solution should be instructive from certain points 
of view. W. E. MILuEr. 
Publication Bureau, General Electric Co., 
Schenectady, New York, U.S.A. 
Tue solid is one of revolution (evidently), and the 
attraction being a maximum is unaltered by shifting a 
small elementary ring of matter from one point to another 
of its bounding surface. If dM is the mass of a ring 
formed by the revolution of the point 7, @, then the attra¢- 
tion is dMcos@/r?. Hence the equation of the generating 
curve of the boundary is cos @/r*=const., or r*=k* cos@ 
say, or (x*?+y*)*=k*x?. The curve may be traced by 
drawing the circle r=kcos@, and taking on each radius 
vector a mean proportional between that radius and k. 
According to this result, the form of the bounding curve 
for a surface of revolution is the same as it would be for 
a plane lamina possessing the same property. The agree- 
ment can be justified by taking a thin slice through the 
axis of the solid. The matter contained in this slice must 
evidently be arranged in such a form as to give the maxi- 
mum attraction independently of the remaining parts of 
the body. G. H. Bryan. 
A New Chemical Test for Strength in Wheat Flour. 
Tue test described as new by Mr. Wood in Nature of 
February 21 has been in use in my laboratory during. the 
past year, where it forms part of the regular routine tests 
applied to flour. While I am fully in agreement with 
Mr. Wood’s view that the volume of carbon dioxide 
evolved by a mixture of yeast and flour under standard 
conditions is a measure of the sugar content of the flour 
together with other fermentable matter produced during 
the fermentative change, it is important not to lose sight 
of the influence exercised by the character of the gluten 
on the volume of the loaf. A rotten gluten when dis- 
tended by too much gas will break, and the gas will 
escape from the dough. From this point of view the 
character of the gluten is clearly of fundamental import- 
ance, but, after all, the problem is one in which no small 
number of variables must be dealt with. 
E. FRANKLAND ARMSTRONG. 
A Remarkable Lunar Halo, February 24. 
In Nature of May 1, 1902 (vol. Ixvi., p. 5), a remark- 
able lunar halo was described as having been witnessed 
from the Yerkes Observatory on January 19, 1902. It con- 
sisted of an ordinary lunar halo, of 45° or 50° in diameter, 
and of a second ring approximately the same in size inter- 
secting the first, and cutting exactly through the moon. 
The same phenomenon was very clearly seen by myself 
and others at Pembroke Dock during the evening of Sunday, 
February 24, between 9 p.m. and 1o p.m. The secondary 
ring appeared to be about a third as large again in 
diameter as the primary, and was situated approximately 
to the north-east of it. In both rings the brownish tinge 
of the edges and dark interiors were perceptible, though 
very much more strongly in the primary than in the 
secondary. 
I should be glad to know whether any explanation has 
yet been advanced as to the optical formation of the 
secondary ring in the above rare phenomenon. 
H. F. Hunt- 
7 Officers Row, Pembroke Dock, Wales, February 26. 
