May 21, 1914| 
of the observatory has been described already (April 
23, p. 201) so it need not be summarised here. 
The complexitv of the relations which research asso- 
ciates and collaborators sustain to the institution is so 
great as to preclude any comprehensive explanation 
within the limits allotted to an annual administrative 
report. Their work embraces a wide range of sub- 
jects, and varies in its conduct from individual inde- 
pendence to intimate collaboration with the depart- 
ments of research and with the division of publica- 
tions. During the past year more than twenty dis- 
tinct fields of research have been cultivated, and a 
total of more than one hundred investigators have 
contributed to the output. Summaries of the work 
of associates proceeding independently are given by 
them in the Year Book. 
THE Bei ECLIPSE Oj gorp In TURKEY 
AND PERSIA. 
N account of the untavourable weather prognos- 
tications for the approaching total eclipse of the 
sun throughout the European countries traversed by 
the tracks of totality, it seems particularly desirable that 
stations should be occupied beyond the Black Sea, 
nearer the sunset limit of eclipse, in eastern Turkey 
and western Persia. 
The central line of the eclipse passes very nearly 
through Baiburt and Bitlis, just a few miles to the 
west of Lake Van, in the former country. In Persia 
it passes through Kermanshah and Khorremabad, to 
the south-west of Teheran, and through a point about 
midway between Persepolis and Dehbid, and slightly 
to the north-east of Shiraz, only a few miles distant 
from Bushire, a port in the north-east of the Persian 
gulf. 
The desert character of a large part of this region 
would indicate that the probability of a cloudless after- 
noon sky in August is very good. Most of this region 
traversed by the shadow is quite elevated, some of it 
being as much as five or six thousand feet above sea- 
level; and this would, in large measure, if not entirely, 
compensate for the lesser altitude of the afternoon sun 
at local totality. Throughout the Turkish region the 
approximate local time of totality is 3h. 50m. p.m., 
the duration of total eclipse being about 120s. 
Throughout the Persian region the time is about 
4h. 50m., with totality shortened to about 105s. As 
very little of the European track has a likelihood of 
less than 50 per cent. of cloud, it seems highly desir- 
able that some of the observers now contemplating 
European location should undertake the extra journey 
into Turkey, at least in order to diminish, if possible, 
the chances of entire failure of the eclipse, such as 
befell astronomers in 1887, and was nearly repeated in 
1896. 
The region of western Persia is not especially diffi- 
cult to reach by way of Batum, at the east end of the 
Black Sea, thence through Tiflis to Baku on the 
Caspian, thence to Resht on the south-west coast of 
the Caspian, whence Kermanshah is easy by caravan 
through Kazbin; or, better, first to Teheran to receive 
Government authority and facilities. _ Most of the 
roads of Persia would permit the use of wheeled 
vehicles only with difficulty. Allow four weeks from 
London or Paris to Teheran, and two weeks thence 
to Kermanshah. Camping outfit and subsistence for 
the most nart should be taken along, as only chicken, 
fruits, and similar edibles can be depended on for the 
last stage of this journey. Roads are in part built, in 
part old roads and trails. From Teheran the best 
route is to Kum, and thence to Sultanabad and Ker- 
manshah; also Bourodjird, quite a large town with a 
telegraph station, and the chief city of Turistan. 
Niggee 325, VOL. 92) 
NATURE 
311 
Summer clouds are said to be highly improbable. 
From Bushire to Shiraz and Persepolis is rather 
more than 100 miles by caravan, the particular draw- 
back at this season being the intense heat, which 
renders travel exceedingly uncomfortable, except at 
night. There are telegraph lines traversing this region 
which would make it feasible for the eclipse observer 
arriving early in the field to check up his longitude 
as well as latitude, so as to make sure of being within 
a few miles of the line of central eclipse. Bushire is 
very accessible ; the steamers of the British India Com- 
pany are scheduled to sail from Bombay every Thurs- 
day; from Karachi every Saturday, and are due in 
Bushire on Wednesday. The Bombay steamers of the 
P. and O. are due to arrive at Bombay on Friday, 
and there is direct rail connection for Karachi, and 
while the British India steamers are scheduled to sail 
from Karachi on Thursday, if the English mails are 
late, the steamers will be held pending their arrival. 
Transportation from Karachi to Bushire is approxi- 
mately 151. 
The Turkish region is very accessible from Tre- 
bizond. The eclipse is total at Trebizond itself, the 
line of exact centrality intersecting the coast a_ few 
miles west of Trebizond, about midway between that 
port and Tereboli. While at the coast towns them- 
selves, including Plattana, Eskiefe, and Jaueboli, the 
chances of clear weather are not at all good, one can, 
by ascending the cliffs and entering the elevated table- 
land of the interior, select observing stations which 
apparently decrease in probable cloudiness, the farther 
inland one goes. Of course, there are no railways ; 
but travelling so far as Erzerum, about 150 miles 
south-east of Trebizond, is not particularly arduous, 
because it is the first section of the early caravan 
route through Tabriz to Teheran. Wheeled vehicles 
are now possible so far as Erzerum, and packages of 
any size and weight required by the eclipse astronomer 
are not prohibited. 
Probably the most detailed map of this region is 
Richard Kiepert’s ‘‘ Karte von Kleinasien,”’ on a scale 
of 1:400,000, published in 1902 by Dietrich Reimer, 
Berlin. The sheets which should be consulted are 
AVI, Tirabzon, and BVI, Erzirum. Another good 
map is the ‘‘Map of Eastern Turkey-in-Asia, Syria, 
and West Persia,’ published by the Royal Geograph- 
ical Society, 1910, and is accompanied by notes. Con- 
sult also ‘‘Zug des Zenophon bis zum Schwarzen 
Meere”’ (Karte ii.), Entworfen von E. v. Hoffmeister, 
accompanying ‘‘Durch Armenien und der Zug Zeno- 
phons” (1911t) and ‘‘Wandkarte des Osmanischen 
Reiches,’”’ von W. v. Diest and Dr. M. Groll (Gea- 
Verlag, Berlin W. 35, 1911); scale 1: 1,250,000. 
Erzerum itself is within the belt of totality, though 
not far from the north-eastern edge of it, so that 
totality would not last more than a very few seconds 
there. Besides this, Erzerum is quite likely to be 
cloudy; and the same might be said of Bitlis itself, 
which is located in a sheltered valley. But about 
fifteen miles west of Bitlis begins the elevated table- 
land of Moush, which, according to the best informa- 
tion I have been able to secure from those resident in 
Bitlis, would probably be cloudless. At the time of 
the eclipse, this whole region rarely experiences any 
rain from the latter part of June until the middle of 
September. The atmosphere is very clear, being only 
a trifle cloudy during that season, and clear skies 
can be depended upon, although it is extremely hot. 
Officers of the Turkish customs are not inclined to 
cause trouble over the baggage of travellers, and it 
is probable that the English and American Consuls 
would be able to get instruments passed without 
examination, especially if the observer brought a letter 
vised by the Turkish Consul nearest his home. 
It would be highly desirable, before leaving home, 
