96 
which I have lately described (Proc. Roy. Soc., 
vol, Ixxxix., p. 133). Moreover, the merging of two 
such series into one formula is open to the objection 
that it involves multiplication by 4 of the series con- 
stant, which would otherwise be universal. It may 
be possible, however, to test this point by observations 
of the Zeeman effects on the lines, and I shall make 
this experiment at the first opportunity. 
I may add that experiments made by Prof. Strutt 
and myself are in harmony with those of Mr. Evans 
in showing that the lines under consideration do not 
occur in mixtures of hydrogen with neon or argon. 
A. FOWLER. 
Imperial College of Science and Technology, 
South Kensington, September 13. 
The Elephant Trench at Dewlish—Was it Dug? 
THE Rev. Osmond Fisher makes the interesting 
suggestion that the curious trough at Dewlish, in 
which numerous remains of Elephas meridionalis were 
found, was an artificial trench, dug as a sort of pit- 
fall to intercept and disable wild animals driven 
across it. Perhaps, as having seen the excavations 
made by Mr. Mansel-Pleydell, I may say a word on 
this point. 
Open trenches in the soft chalk are unknown else- 
where, though they are common enough in the hard 
mountain limestone. I therefore examined this 
trench most carefully, in order to find out how it had 
originated, and whether man had had anything to do 
with it. I am still much puzzled as to its exact mode 
of excavation; but certain peculiarities convinced me 
that it was due to natural agencies, and that it was 
probably cut by the swirl of the fine dust-like quartz- 
sand which, mixed with polished flints, now fills its 
lower part. I could find no implements, and could 
nowhere see traces of pick marks. The sides of the 
trench, where not damaged by the workmen who had 
just cleared it, were curiously smooth; but the flint- 
nodules projected into the cavity from either side, as 
though the softer chalk had been scoured away. The 
abrupt rounded end of the trench was most peculiar, 
and as I cleaned this out myself, dusting away the 
sand from the smoothed face of the chalk, I am sure 
that there were here neither tool-marks nor rubbings 
such as might be made by a man working in the 
trench, or by wild beasts. In short, the smooth, 
rounded contours suggested the eddying of wind, and 
the absence of any crack or joint showed that here 
at any rate the rounding was not likely to be due to 
percolating water. 
Beneath the elephant bones, which occurred in a 
layer a few feet down, the infilling of the trench 
seems to be a fine dust-like, unfossiliferous sand, 
which was not bottomed, as Mr. Mansel-Pleydell’s 
excavations were made primarily to obtain elephant 
remains, and these were in such a soft condition as 
to make removal almost impossible. If this sand- 
filled fissure is found to continue downward, but is 
too narrow for a man to work in, it will show that 
the trench is not artificial. I could only just squeeze 
past in one or two places; but the upper part of the 
trench was passable; I think, however, that it tended 
to narrow downward, but at the time of my visit the 
bones had not been removed, and I could not excavate 
below them. 
Perhaps someone acquainted with plateaus of soft 
limestone under desert conditions can say whether 
there is any tendency for the wind to cut trenches 
with rounded blind ends, such as the Dewlish trench 
has. In this connection, it is worth noting that our 
newer Pliocene Jand-faunas show distinct indications 
of drier and more sunny conditions than we have at 
NO. 2291, VOL. 92] 
NATURE 
[SEPTEMBER 25, 1913 
present. A gazelle, an antelope, and several land and 
fresh-water mollusca point in that direction. Under 
_ dry conditions, and before the loose flints were swept 
away during the glacial period, our chalk-downs 
' would probably be stony deserts, quite unlike the 
green hills we now see. CLEMENT REID. 
Milford-on-Sea. 
Red-water Phenomenon due to Euglena. 
THE red-water phenomenon due to a Euglena - 
described by Prof. Dendy in Nature of August 7 has 
been observed by me in Pretoria. In this case, how- 
ever, the Euglena swims freely about in the water, 
and also forms a red gelatinous scum on the surface 
of the damp mud on the side of the pond. In swim- 
ming they seldom show euglenoid. movement. A 
flagellum longer than the body can be easily seen 
under the microscope at the anterior end of the body, 
but it always trails along the body with lashing 
movements. If they become stranded on the mud at 
the edge of the pond, they soon become spherical and 
encysted in a mucilaginous covering much wider than 
the body and showing a layered formation. I have 
not observed any bubbles of gas given off, although | 
have kept large quantities of them under observation 
for long periods. They appear to prefer the encysted 
form, as they always swim to the edge of the vessel 
towards the light and form a deep red line along the 
edge, which gradually becomes dry. If more water 
is added and the vessel turned round, they will leave 
their cysts and again swim towards the light side. 
They are of a fairly large size, and have a cylindrical 
body tapering to a sharp point at the posterior end, 
where the last portion is free from pigment. Chloro- 
phyll is present, and is easily seen amongst the red 
in those that have just come out of the encysted 
stage, but later on it entirely disappears. ‘ 
Horace A. WaGER. 
Transvaal University College, Pretoria. 
August 30. 
Distance of the Visible Horizon. 
Mr. W. Moss’s account in Nature for August 7, 
583, as to how to get the area of a sphere 
theoretically visible at any altitude is interesting; but 
can he, or any of your readers, say what the formula 
is for obtaining the distance actually visible with 
an average amount of refraction? So far as I can 
discover, all ordinary books of tables ignore this, 
although such a table would be very useful. 
A table is given in Chamber’s Mathematical Tables, 
p. 436, for the distance of the visible horizon, but the 
explanation, p.-xl., states that this is theoretical, and 
that a correction for refraction should be made, 
although nowhere is any table or formula given for 
such correction. T. W. BackHOUSE. 
West Hendon House, Sunderland. 
September 6, 1913. 
ATMosPHERIC refraction is such a varying quantity 
that no rule respecting it can be laid down applicable 
in all circumstances; as in cases of mirage, for 
instance, where vessels below the horizon are seen 
standing above it, and turned upside down. The 
refraction of the sea horizon is the great difficulty in 
obtaining correctly the position of vessels at sea. 
This can be eliminated in most cases by taking 
observations of the heavenly bodies to opposite sides 
of the horizon; for latitude in a north as well as 
in a south direction; for longitude in an east as 
well as in a west direction. When only one heavenly 
object is available this is not always practicable, but 
it can be done when the altitude is 60° or upwards. 
i i ee ee i 
a 
