SEPTEMBER 14, 1916| 
NATURE 
41 
the 30-ft., and the first order of the 75-ft. spectro- 
graph, and in the case of the closest pairs curves were 
also obtained with the registering photomicrometer ; 
the separations derived by the latter process appeared 
to be least affected by accidental or systematic errors. 
three methods were systematically smaller than those 
indicated by Rowland’s tables, the difference varying 
inversely as the separation. For six pairs, mean 
separation 0-274, the difference was +0-003; for eight 
pairs, mean separation 0-145, it was + 0-008; and for 
eight other pairs, mean separation 0-075, it was 
+0:013. These differences are regarded as being prob- 
ably due to errors in the Rowland values. The filar- 
micrometer values were found to vary with the width 
of the slit and the density of the spectrograms; what- 
ever decreased the intensity of the space between the 
two components, as compared with the continuous spec- 
trum outside, led to over-estimation of the interval. 
These results have evidently an important bearing 
upon recent attempts to detect effects due to anomalous 
dispersion. 
. The curious personal errors in the measurement of | 
close lines have also been discussed from another point 
of view by H. H. Plaskett, in a paper entitled ‘‘ The 
Psychology of Differential Measurements”’ (Journal 
R.A.S. Canada, June, 1916; Nature, vol. xcvii., 
Pp. 451.) 
» SPECTRA AND ABSOLUTE MaGNITUDES OF StTARS.—It 
has previously been shown by Adams that stars of 
small proper motion are relatively weaker in the more 
refrangible parts of the spectrum than stars of large 
proper motion, and that for stars of classes F to K 
this difference increases with advancing type. While 
the first result suggests a scattering of light in space, 
the second indicates that the absorption in the violet 
depends, in part at least, on the physical state of the 
star. This investigation has recently been extended at 
Mount Wilson by G. S. Monk, who has examined 
about 1200 plates of stellar spectra (Astrophysical 
Journal, vol. xliv., p. 45). The results are in general 
agreement with those of Adams, but show the effects 
to a less extent. All the density measures which 
could be so used were further discussed in relation to 
the absolute magnitudes of the stars, as determined 
by Adams on the basis of intensities of certain special 
lines. A relationship was thus indicated as existing 
between absolute masnitude and the relative weakness 
in the violet part of the spectra of stars having small] 
proper motions, and it is concluded that the greater 
part of this effect is not due to absorption of light in 
space. It is thought probable that, with the aid of 
photographs specially taken for the purpose, the rela- 
tive intensity of the violet part of the spectrum, to- 
gether with spectral type, might be successfully em- 
ployed to provide fairly accurate values of absolute 
magnitudes. An additional spectroscopic method of 
determining stellar distances is thus suggested. 
OBSERVATIONS OF MINOR PLANETS IN FRANCE.—It is 
gratifying to find that in spite of difficulties caused by 
the war, the French observatories have been able to 
secure a large number of observations of minor planets. 
A useful summary of these observations, made during 
the year 1915, is given in the Journal des Observateurs, 
vol. i., No. ro. Although no new discoveries were 
made by French astronomers, an abundance of precise 
observations of about 120 known minor planets were 
obtained. About two-thirds of the observations were 
made at Algiers, on plates taken with the instrument 
employed for the photographic chart of the heavens. 
In addition, numerous ephemerides, in many cases 
constructed from corrected orbital elements, were 
issued by the Marseilles Observatory. 
NO. 2446, VoL. 98] 
MARITZBURG MEETING OF THE 
SOUTH AFRICAN ASSOCIATION. 
-1>HE fourteenth annual session of the South African 
Association for, the Advancement of Science was 
THE 
The mean separations determined at Mount Wilson by | held in Maritzburg, the capital of the Province of 
Natal, on July 3-8, under the presidency of Dr. 
Lawrence Crawford, professor of mathematics in. the 
South African College, Capetown. The meetings of 
the sections took place in what used to be the Natal 
Houses of Parliament, the buildings of which are now 
the abode of the Natal Provincial Council. On the 
afternoon of the opening day of the session the presi- 
dent and members of the association were officially 
welcomed by the Mayor and Corporation of Maritz- 
burg, and in the evening his worship gave a recep- 
tion in the Town Hall, after which the president took 
the chair and delivered his address. 
About eighty-five papers were read in the four sec- 
tions into which the association is usually divided, 
and summaries of a few of these are given below. 
Prof. John Orr, who will preside over the 1917 
session of the association, which is to be held at Stellen- 
bosch, Cape Province, was president of Section A at 
Maritzburg, and his presidential address, which was 
profusely illustrated by means of lantern slides, was 
largely a review of the progress of engineering science 
in South Africa of late years, particularly in connec- 
tion with mining operations on the Rand. 
The presidential address in Section B was given by 
Prof. J. A. Wilkinson, professor of chemistry in the 
South African School of Mines and Technology, 
Johannesburg. He laid stress on the fact that South 
| Africa continues to exist on its rich stock of raw 
materials—its exports, in addition to the raw pro- 
ducts of agriculture, being chiefly metals, crude and 
unrefined, and uncut diamonds. He devoted his ad- 
dress to urging the need of organising the develop- 
ment of chemical industry and research in the widest 
sense of those terms. He deplored the existence of 
the popular impression that the duties of the chemist 
and the pharmacist are identical, and to this cause 
he ascribed the fact that South African products were 
not up to the standard of imported goods. In con- 
clusion, he submitted details of a scheme for organis- 
ing chemical research in South Africa upon sound 
lines, and urged that this organisation should be 
undertaken without delay. 
Mr. I. B. Pole Evans, chief of the division of botany 
in the Union Department of Agriculture, was presi- 
| dent of Section C. His presidential address took the 
form of a sketch of the rise and development of myco- 
logy in South Africa. Persoon, the father of the 
science of mycology, he pointed out, was a South 
African, having been born at the Cape in 1755, and 
some of the earliest references to Cape fungi are in 
his ‘‘Synopsis methodica fungorum,” published in 
1801.- During. the: course of his address Mr. Pole 
Evans: incidentally referred to his own appointment 
as mycologist to the Transvaal Government in 1905. 
He could not be expected to know that some years 
previously the Cape Government, too, had put a sum 
of money: on the: Estimates for the salary of a myco- 
logist, but so much sport was made of the item in 
the Cape Parliament that it was hurriedly withdrawn. 
“The ignorance of the subject and the vague notions 
that people have about fungi are,’’ said Mr. Pole 
Evans, ‘‘due chiefly to the fact that most of them are 
microscopic, and consequently cannot easily be con- 
ceived by those unfamiliar with the life of the unseen 
world.”’ : 
Section D- was: presided over by Mr. M. S. Evans, 
who took-as the subject of his address a-survey of the 
past and present relations of the European and Bantu 
