186 TROWBRIDGE. — IIIGII ELECTROMOTIVE FORCE. 
* 
each cell is approximately 2 volts and the internal resistance 1.50 volts; the total elec- 
tromotive force of the battery is 40,000 volts. It can produce in air an initial spark 
of 3.5 cm. to 4 cm., and when the terminals are quickly drawn apart, a flaming dis- 
charge of 60 cm. 
This modern large installation has been made possible by the discovery of the Plante 
accumulator, which enables one to utilize the powerful currents generated by the 
dynamo machine to polarize lead plates and 
dispen 
Th 
latter have entirely disappeared from physical laboratories except for minor purposes, 
such as call bells; and the three or four hundred Bunsen or Grove cells which were 
once used in lectures on physics have been thrown into the scrap heap. 
I published in the Proceedings of the American Academy, Vol. XXXII, p. 253, 
also Vol. XXXIII, p. 435, a description of this battery, which I had constructed for the 
purpose of studying the spectra of hydrogen, especially 
f hyd 
lines discovered by Professor E. C. Pickering in the spectrum of certain stars. It was 
hoped that under conditions of high electromotive force and strong currents this new 
series might be reproduced in a laboratory. Although this hope was found to be falla- 
cious, the installation of such a large plant has proved extremely useful where a steady 
electric field is necessary. 1 It has shown that spectrum analysis of rarefied gases is 
practically limited to comparatively small electromotive forces and weak currents. An 
experience of twelve years in the use of the battery impels me to gather in a memoir 
the results of my 
The life of the battery as it was originally constructed 
has been ten years, and the improvements I shall describe in this memoir will, I believe, 
greatly extend its life. 
The form of cells which I adopted, consisting of test tubes enclosing strips of lead 
formed by the Plante method, is undoubtedly the cheapest and most compact form 
that can be selected. A less number of cells, of 
might be 
far more 
o 
d of the pasted type 
•- 
but they would occupy a much greater space and would be 
The battery with which I have experimented requires a room 20 x 30 x 14 feet. 
It is contained in eight cabinets or cases. The space occupied by the battery could 
be lessened by a more compact arrangement of the trays which hold the cells. 
Each of the unit trays consists of three compartments (Figure 1, Plate XXV), and 
holds sixty cells. The supports of the test tubes are made of kiln-dried whitewood 
soaked in paraffine. The tubes rest on a lower shelf, and are held upright in holes in 
an upper shelf. The individual trays, comprising together what I have termed a unit, 
1 II. L. Blackwell, Proc. Am. Acad., Vol. XLT, No. 32. 
