—— 
May 27, 1886] 
NATURE 
is 
Scientific Memoirs by Medical Officers of the Army of 
India. "Edited by B. Simpson, M.D., Surgeon-General 
with the Government of India. (Part I., 1884). (1) 
On the relation of cholera to Schyzomycete organisms, 
by D. D. Cunningham. (2) On the presence of peculiar 
parasitic organisms in the tissue of a specimen of Delhi 
boil, by D. D. Cunningham. (Calcutta, 1885). 
In the first of these memoirs Dr. Cunningham makes 
some interesting additions to our knowledge of the 
presence and distribution of comma bacilli in the in- 
testinal contents in cases of Asiatic cholera; on the 
occurrence of peculiar comma bacilli associated with the 
scum formed on tank water by Euglenz ; and on certain 
modifications in morphological and other characters in 
artificial cultivations of the choleraic comma bacilli. 
The second memoir gives a minute description of the 
anatomical nature of the skin disease known as ‘‘ Oriental 
sore” or “Delhi boil.” This description is the more 
valuable as it is the first accurate account that we possess 
of the minute anatomy of this interesting malady. The 
value is enhanced by the discovery by Dr. Cunningham in 
the diseased tissue of a peculiar fungus bearing the 
characters of Mycetozoa or Myxomycetes, more especially 
of the subdivision of the Monadine ; the distribution of 
this fungus is such that a causal relation of it to the 
disease process becomes highly probable. 
The memorr is illustrated by numerous fine lithographs, 
many of them coloured. E. KLEIN 
LEDLIERS LO THE EDITOR. 
[The Editor does not hold himself responsible for opinions ex- 
pressed by his correspondents. Neither can he undertake to 
return, or to correspond with the writers of, rejected manu- 
scripts. No notice is taken of anonymous communications. 
[The Editor urgently requests correspondents to keep their letters 
as short as possible. The pressure on his space is so great 
that it ts impossible otherwise to insure the appearance even 
of communications containing interesting and novel facts.] 
On the Thomson Effect as Expounded by Prof. Tait 
AMONG modern expositions of the subject of thermo-electri- 
city there is none so full, and on the whole so instructive to 
students, as that contained in Prof. Tait’s ‘‘ Heat.” It is there- 
fore the more important to call attention to what appears to me 
to be, to say the least, a very questionable statement there made. 
It refers to the Thomson effect. 
Thomson’s experiments were of the following nature. A 
metallic bar was surrounded with a hot-water jacket in the 
middle and with cold-water jackets at the ends, and there were 
two holes sunk in it for the insertion of thermometers, midway 
between the hot jacket and the two cold jackets. When the 
flow of heat had become nearly steady, a steady current of 
electricity was sent through the bar ; and, after it had flowed 
for several minutes in one direction, it was reversed ; then, after 
the same number of minutes, it was again reversed, and so on 
several times. It was thus found that, when the bar was of 
copper, the current made the temperature of the further ther- 
mometer higher than that of the near one (the words fay and 
near being used with reference to the end at which the current 
entered). When the bar was of iron, the current made the 
temperature of the near thermometer higher than that of the 
further one. f 
Seeing that a current may be regarded at pleasure as the flow 
of vitreous electricity in the nominal direction of the current, or 
as the flow of resinous electricity in the opposite direction, 
Thomson summed up his results by saying that ‘‘ the vitreous 
electricity carries heat with it” in copper, and ‘‘the resinous 
electricity carries heat with it” in iron. He also gave the name 
of ‘electric convection of heat” to the effect thus detected. 
It has since been called by others ‘‘the Thomson effect.” 
The experiments were instituted to test the truth of a conclu- 
clusion of which he had previously given a theoretical proof— 
the conclusion that ‘‘in one or other of the metals, and most 
probably in both, there must be a thermal effect due to the 
passage of electricity through a non-uniformly heated portion of 
it, which must be an absorption of heat [a cooling] or an evolu- 
tion of heat [a warming], according to the direction of the 
current between the hot and cold parts.” 
It may be taken to be an established fact that, in a uniform 
linear conductor along which a current is flowing, there is, in 
addition to the frictional heating, which is proportional to the 
square of the current, a warming or cooling effect proportional 
(at given temperature) to the steepness of the thermometric 
gradient at the point which is warmed or cooled, changing sign 
with the gradient, and vanishing at points of maximum or 
minimum temperature, where the gradient vanishes. 
Now compare these effects with what happens when a stream 
of liquid flows through a pipe surrounded at alternate points in 
its length with hot and cold jackets, the average temperature of 
the water being the same as the average temperature of the 
pipe. It will carry heat from the hotter to the colder portions, 
thus cooling the hottest parts, warming the coldest parts, and at 
the same time carrying forward the points of maximum and 
minimum temperature. If, at each point of the pipe (supposed 
straight and horizontal), we erect an ordinate to represent its 
temperature, and call the curve of which they are the ordinates 
“the temperature curve,” the effect of the flow of liquid on this 
curve will be twofold: (1) it will carry the temperature curve 
forward ; (2) it will make the temperature curve flatter. 
Thomson’s experiments show that an electric current carries 
the temperature curve forward in copper, and backward in iron ; 
but I am not aware of any evidence to show that it makes the 
temperature curve flatter. 
The analogy between the Thomson effect and convection of 
heat by a liquid ina pipe therefore does not run on all fours, 
and must be used with caution. 
Maxwell says (‘‘Elec. and Mag.,” p. 343, second edition), 
** positive electricity in copper, and negative electricity in iron, 
carry heat with them from hot to cold.” The words “‘ from hot 
to cold” are here added to Thomson’s original phrase ‘‘ carries 
heat with it,” and the addition thus made is not in accordance 
with facts, for it implies that heat is taken away from the hot 
parts and given to the cold parts ; whereas the fact is that heat 
is taken from parts where the temperature gradient is in 
one direction, and heat is given to parts where the gradient is 
in the opposite direction. If the statement be altered by a little 
transposition, so as to make it stand thus, ‘positive electricity 
.in copper, and negative electricity in iron, going from hot to 
cold, carry heat with them,” it will be scarcely distinguishable 
from Thomson’s original statement. 
Prof. Tait goes further, and says (‘‘ Heat,” p. 170) :—“‘ After 
a series of elaborate experiments (described in the Phi. Trans. 
for 1855) [it should be 1856] Thomson found that :— 
“« 4n electric current in an unequally heated copper conductor 
behaves as a real fluid would do, i.e. it tends to reduce differenzes 
of temperature. In iron it tends to exaggerate them.” 
The italics are Prof. Tait’s. 
I can find nothing in Thomson’s paper to support the asser- 
tion that in copper an electric current tends to reduce differ- 
ences of temperature, though the idea that it does so is natur- 
ally suggested by the analogy implied in the phrase “‘ electrical 
convection of heat.” 
The statement that in iron the current tends to exaggerate 
differences of temperature, seems to be completely original on 
the part of Prof. Tait. It does not arise naturally out of 
Thomson’s dictum, ‘‘ resinous electricity carries heat with it in 
iron”; for if we think of resinous electricity as a real fluid 
flowing through iron, it would tend to equalise differences of 
temperature in that metal. 
The two statements taken together suggest the following line 
of reasoning as conclusive against them both :— 
Let there be the same initial distribution of temperature in a 
copper and in an iron bar, and currents in the same direction 
through both. Then the alterations of temperature at corre- 
sponding points in the two bars will have opposite signs, Any 
one who maintains that the warmest parts of the copper are 
cooled is therefore bound to maintain that the warmest parts of 
the iron are warmed. But there is precisely the same ground 
for maintaining that the warmest parts of the iron are cooled, 
and therefore the warmest parts of the copper warmed. What- 
ever vitreous electicity can do in copper, resinous electricity can 
do in iron. We are thus involved in a contradiction if we 
assume any finite heating or cooling at the hottest parts. And 
similar reasoning disproves any finite heating or cooling at the 
coldest parts. 
The following formal investigation confirms the view which I 
have above expressed. 
