
JULY 2, 1915] 
NATURE 
561 

LETTERS TO THE EDITOR. 
[The Editor does not hold himself responsible for 
opinions expressed 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 cnonymous communications. | 
The Structure of Magnetite and the Spinels, 
THE structure of the spinel group of crystals is very 
interesting. These crystals are cubic, and possess the 
greatest possible number of symmetries. The com- 
position is given by the formula, R’R’”,O,, where the 
divalent metal R’’ may be Mg, Fe, Zn, or Mn, and 
the trivalent metal R’” may be Fe, Mn, Er or Al: 
Magnetite is FeFe.O,. 
‘The structure is fundamentally the same as that of 
the diamond. Each carbon atom of the diamond is 
to be replaced by the divalent metal atom; the distance 
between two neighbours being 3.60 A.U. in magnetite 
as against 1-53 A.U. in diamond. The four oxygen 
atoms are arranged in a regular tetrahedron about 
the divalent atom. The lines joining the latter to the 
former are parallel to the four cube diagonals. Any 
two neighbouring tetrahedra point towards each 
other. If each perpendicular from a_ tetrahedron 
corner on the opposite face is produced it encounters 
another tetrahedron, passing’ first through the middle 
point of a face and then through the opposite corner. 
A trivalent atom lies on each such connecting line, 
half-way between the tetrahedra.. The distance be~ 
tween a divalent and the nearest trivalent atom is 
720 A.U. Four trivalent atoms are associated with 
each tetrahedron, but each atom is shared by two 
tetrahedra. As in other cases already examined, the 
molecule has no separate existence. The size of the 
tetrahedron may not be the same in all members of the 
group of crystals. The divalent atom lies at the centre 
of a tetrahedron of oxygen atoms, and the trivalent 
at the centre of an octahedron. 
W. H. Brace. 
Leeds, July 12. 

The Magnetic Storm of June 17, and Aurora. 
Pror. BaRNarD’s interesting letter dated June 25, 
in Nature of July 15, on what is termed “The Great 
Aurora of June 16, 1915,” is at first sight rather 
puzzling to the non-astronomical reader. The large 
magnetic storm began about 1.50 a.m. on June 17. 
On June 16, it is true, there was a magnetic disturb- 
ance, but not such as to suggest a striking auroral 
display. The explanation presumably is that Prof. 
Barnard is referring to an astronomical day, com- 
mencing at Greenwich noon on June 16. This, at 
least, would explain his statement that at Wisconsin 
(about 90° W.) at 2th. 25m. “the sky was bright with 
dawn.’’ This one would expect between 3 and 4 a.m. 
local time. If this is correct, then the first auroral ap- 
pearance chronicled by Prof. Barnard was at 3.30 a.m. 
on June 17, Greenwich civil time, and the maximum 
brilliancy about 8.15 a.m. It was principally during 
these morning hours that the Kew magnetic curves 
had the rapid oscillatory character usually associated 
with aurora and earth currents. The newspaper re- 
ports quoted by Prof. Barnard seem to fit this explana- 
tion. . 
Passing to the Rev. A. L. Cortie’s letter (p. 537), it 
really emphasises the difficulty of deciding whether 
individual sun-spots and magnetic storms are con- 
nected. There are often a number of spots visible at 
one time. A spot remains visible for a number of 
days, during which there may be several magnetic 
storms. If spots cause storms, the rule one spot one 
storm may not be observed. If I selected any given 
NO. 2386, VOL. 95 | 
| for 

date, storm or no storm, the chances are Father 
Cortie could supply a spot. I think Father Cortie 
has not quite grasped my argument that quiet days 
show the twenty-seven-day period equally with dis- 
; turbed days, and that one can scarcely associate them 
with limited areas or “‘anti-spots.”” if one associates 
them, as he now seems to do, with an undisturbed 
state of a whole solar hemisphere, why not equally 
associate storms with a generally disturbed state of a 
whole hemisphere? As a matter of fact, the average 
quiet day seems associated with a practically average 
state of solar spottedness. The 600 quiet days selected 
by the Astronomer Royal from 1890 to 1g00 gave 
Wolfer’s provisional sun-spot frequency a mean 
value of 41-28, the mean from all days of the eleven 
years being 41-22. They showed the twenty-seven-day 
period very clearly. C. CHREE. 
Richmond, Surrey, July 17. 
Surface Tension and Ferment Action. 
In Nature of June 17 Messrs. E. F. and H. E. 
Armstrong criticise the conclusions drawn by Mr. 
Beard and myself in a paper published in the Proc. 
Roy. Soc. of June 1, under the title ‘‘ Surface Tension 
and Ferment Action.” We drew the conclusion that 
the action of invertase was inhibited by surface ten- 
sion. According to Messrs. Armstrong the inhibition 
observed under the conditions of our experiments was 
due simply to a minute trace of alkali given off by 
the glass. They state in confirmation of their view 
that the action of the alkali given off by ordinary 
glass is so marked that it is impossible to obtain 
consistent results with invertase, unless hard glass 

vessels, test-tubes, and storage bottles are used. That 
is certainly not our experience. We failed to find any 
difference in the readings between two mixtures of 
cane-sugar and invertase, of which one was kept in 
contact with glass beads at medium temperatures, as 
long as the amount of invertase used was relatively 
large. Our experience in that respect is apparently 
in accordance with that of Sdrensen, who states that 
the effect of the alkalinity of glass makes itself felt 
only in the case of invertase solutions which have 
been especially purified. 
In our experiments an inhibition was noticed only 
when the amount of invertase was relatively small. 
Under these conditions an alteration in the hydrogen- 
ion concentration produced by the minute trace of 
alkali given off by glass may have had some share in 
producing an inhibition, but it does not account for 
certain features of the phenomenon, which we have 
been careful to emphasise in our paper. If the alkali 
from the glass was entirely responsible for the effect 
one would expect the inhibition to persist in its entirety 
after the glass beads have been removed. This was 
found not to be the case. Again, the weakening of 
an invertase solution, which had been allowed to stand 
in contact with glass beads at medium temperatures 
and which we ascribed to absorption of the ferment 
by glass, cannot be explained on the ground put for- 
ward by Messrs. Armstrong. Their view necessitates 
the assumption of so large an amount of alkali given 
off by glass to an invertase solution, that it should be 
detectable by such an indicator as phenolphthalein. 
This again was not the case. 
The interruption of my work has unfortunately 
delayed the completion and publication of similar 
observations with diastase carried out by Mr. McCall 
and myself. It was found that the inhibition pro- 
duced by extending the surface-glass water could be 
almost completely removed by coating the glass with a 
thin film of a surface-active substance, such as methyl 
alcohol, ethyl alcohol, amyl alcohol, ether. On the 
other hand, films of ligroin and xylol deposited on the 
glass failed to remove the inhibition. 
