the south, 2 in the south-east, 4 in the east, 2 in the central, and 2 in the north-east. It is noteworthy that, in all these sub-regions save the south-east, turacin-bearers are found along with those plantain-eaters which do not contain the pigment. Oddly enough two of the latter species, Schizorhis africana and S. zonura, possess white patches destitute of pigment in those parts of the feathers which in the turacin-bearers are crimson. These birds do not, I will not say cannot, decorate these bare patches with this curiously complex pigment. [Some extracts were here given from the late Mr. Monteiro's book on Angola, vol. ii. pp. 74–79, and from letters by Dr. B. Hinde. These extracts contained references to curious traits of the touracos.]

Usually from 12 to 18 of the primaries or metacarpo-digitals and secondaries or cubitals amongst the wing feathers of the turacinbearers have the crimson patches in their web. Occasionally the crimson patches are limited to six or seven of the eleven primaries. I have observed this particularly with the violet plantain-eater (Musophaga violacea). In these cases the crimson head-feathers, which also owe their colour to turacin, are few in number, as if the bird, otherwise healthy, had been unable to manufacture a sufficiency of the pigment. I may here add that the red tips of the crest feathers of Turacus meriani also contain turacin.

In all the birds in which turacin occurs, this pigment is strictly confined to the red parts of the web, and is there unaccompanied by any other colouring matter. It is therefore found that if a single barb from a feather be analysed its black base and its black termination possess no copper, while the intermediate portion gives the bluegreen flash of copper when incinerated in the Bunsen flame. [A parti-coloured feather was burnt in the Bunsen flame, with the result indicated.]

Where it occurs, turacin is homogeneously distributed in the barbs, barbicels and crochets of the web, and is not found in granules or corpuscles.

In

To the natural question "Does turacin occur in any other birds besides the touracos?" a negative answer must at present be given. At least my search for this pigment in scores of birds more or less nearly related to the Musophagidæ has met with no success. some of the plantain-eaters (species of Turacus and Gallirex) there is, however, a second pigment closely related to turacin. It is of a dull grass-green colour, and was named Turacoverdin by Dr. Krukenberg in 1881. I had obtained this pigment in 1868 by boiling turacin with a solution of caustic soda, and had figured its characteristic absorption band in my first paper (Phil. Trans., vol. clix. 1870, p. 630, fig. 4). My product was, however, mixed with unaltered turacin. But Dr. Krnkenberg obtained what certainly seems to be the same pigment from the green feathers of Turacus corythaix, by treating them with a 2 per cent. solution of caustic soda. I find, however, that a solution of this strength

dissolves, even in the cold, not only a brown pigment associated with turacoverdin, but ultimately the whole substance of the web. By using a much weaker solution of alkali (1 part to a thousand of water) a far better result is obtained. [The characteristic absorption band of turacoverdin, which lies on the less refrangible side of D, was shown; also the absorption bands of various preparations of turacin.] I have refrained from the further investigation of turacoverdin, hoping that Dr. Krukenberg would complete his study of it. At present I can only express my opinion that it is identical with the green pigment into which turacin when moist is converted by long exposure to the air or by ebullition with soda, and which seems to be present in traces in all preparations of isolated turacin however carefully prepared.

A few observations may now be introduced on the physical and chemical characters of turacin. It is a colloid of colloids. And it enjoys in a high degree one of the peculiar properties of colloids, that of retaining, when freshly precipitated, an immense proportion of water. Consequently, when its solution in ammonia is precipitated by an acid, the coagulum formed is very voluminous. [The experiment was shown.] One gram of turacin is capable of forming a semi-solid mass with 600 grams of water. Another character which turacin shares with many other colloids is its solubility in pure water and its insolubility in the presence of mere traces of saline matter. It would be tedious to enumerate all the observed properties of turacin, but its deportment on being heated and the action of sulphuric acid upon it demand particular attention.

At 100° C., and at considerably higher temperatures, turacin suffers no change. When, however, it is heated to the boiling-point of mercury it is wholly altered. No vapours are evolved, but the substance becomes black and is no longer soluble in alkaline liquids, nor, when still more strongly heated afterwards, can it be made to yield the purple vapours which unchanged turacin gives off under the same circumstances. This peculiarity of turacin caused great difficulty in its analysis, for these purple vapours contain an organic crystalline compound in which both nitrogen and copper are present, and which resist further decomposition by heat. [Turacin was so heated as to show its purple vapours, and also the green flame with which they burn.] This production of a volatile organic compound of copper is perhaps comparable with the formation of nickeland ferro-carbonyl.

The action of concentrated sulphuric acid upon turacin presents some remarkable features. The pigment dissolves with a fine crimson colour, and yields a new compound, the spectrum of which presents a very close resemblance to that of hæmatoporphyrin [Turacin was dissolved in oil of vitriol: the spectrum of an ammoniacal solution of the turacoporphyrin thus produced was also shown], the product obtained by the same treatment from hæmatin: in other respects also this new derivative of turacin, which I call turacoporphyrin, reminds one

of hæmatoporphyrin. But, unlike this derivative of hæmatin, it seems to retain some of its metallic constituent. The analogy between the two bodies cannot be very close, for if they were so nearly related as might be argued from the spectral observations, hæmatin ought to contain not more but less metal than is found to be present therein.

The percentage composition of turacin is probably-Carbon 53.69, hydrogen 4.6, copper 7.01, nitrogen 6 96, and oxygen 27.74. These numbers correspond pretty nearly to the empirical formula, C2 H81 Cu, N, O32. But I lay no stress upon this expression.

82

I have before said that copper is very widely distributed in the Animal Kingdom. Dr. Giunti, of Naples, largely extended (1881) our knowledge on this point. I can hardly doubt that this metal will be found in traces in all animals. But besides turacin only one organic copper compound has been as yet recognised in animals. This is a respiratory, and not a mere decorative, pigment like turacin. Léon Fredericq discovered this substance, called hæmocyanin. It has been observed in several genera of Crustacea, Arachnida, Gastropoda and Cephalopoda. I do not think it has ever been obtained in a state of purity, and I cannot accept for it the fantastic formula-C867 H1369 Cu SO259-which has recently been assigned to it. On the other hand, I do not sympathise with the doubts as to its nature which F. Heim has recently formulated in the Comptes Rendus.

It is noteworthy, in connection with the periodic law, that all the essential elements of animal and vegetable organic compounds have rather low atomic weights, iron, manganese and copper representing the superior limit. Perhaps natural organic compounds containing manganese will some day be isolated, but at present such bodies are limited to a few containing iron, and to two, hæmocyanin and turacin, of which copper forms an essential part.

66

If I have not yet unravelled the whole mystery of the occurrence and properties of this strange pigment, it must be remembered that it is very rare and costly, and withal difficult to prepare in a state of assured purity. It belongs, moreover, to a class of bodies which my late master, Dr. A. W. von Hofmann, quaintly designated as dirts" (a magnificent dirt truly!)-substances which refuse to crystallize and cannot be distilled. I have experienced likewise, during the course of this investigation, frequent reminders of another definition propounded by the same great chemist, when he described organic research as "a more or less circuitous route to the sink!"

I am very glad to have had the opportunity of sharing with an audience in this Institution the few glimpses I have caught from time to time during the progress of a tedious and still incomplete research into the nature of a pigment which presents physiological and chemical problems of high if not of unique interest.

Let my last word be a word of thanks. I am indebted to several friends for aid in this investigation, and particularly.to Dr. MacMunn, of Wolverhampton, the recognised expert in the spectroscopy of animal pigments. [À. H. C.].

VOL. XIV. (No. 87.)

E

WEEKLY EVENING MEETING,

Friday, February 24, 1893.

SIR FREDERICK ABEL, K.C.B. D.C.L. D.Sc. F.R.S. Vice-President, in the Chair.

EDWARD HOPKINSON, Esq. M.A. D.Sc.

Electrical Railways.

ONE of the most striking of the many new departures in the practical application of electrical science, which made the Paris Exhibition of 1881 memorable, was a short tramway laid down under the direction of the late Sir William Siemens, from the Palais de l'Industrie to the Place de la Concorde, upon which a tramcar worked by an electric motor plied up and down with great regularity and success during the period of the Exhibition. Yet few of those who saw in this experiment the possibilities of a great future for a new mode of traction would have ventured to predict that within ten years' time, in the United States alone, over 5000 electric cars would be in operation, travelling 50,000,000 miles annually, and carrying 250,000,000 passengers, or that electrical traction would have solved the problem of better communication in London and other large cities. Two years before the Exhibition in Paris the late Dr. Werner Siemens had exhibited at the Berlin Exhibition in 1879 an experimental electric tramway on a much smaller scale, and his firm had put down in 1881 the first permanent electric railway in the short length of line at Lichterfelde, near Berlin, which, I believe, is still at work. In the same year Dr. William Siemens undertook to work the tramway, then projected, between Portrush and Bushmills, in the North of Ireland, over six miles in length, by electric power, making use of the waterpower of the Bush River for the purpose, an undertaking which I had the advantage of carrying out under his direction. It is no part of my object to-night to follow further the history of electric traction, which is so recent that it is familiar to all; but, in alluding to these initial stages of its development, I have desired to recall that it was the foresight and energy of Dr. Werner and Dr. William Siemens, and their skill in applying scientific knowledge to the uses of daily life, which gave the first impulse to the development of the new electrical power.

The problem of electric traction may be naturally considered under three heads :

(1) The production of the electrical power.

(2) Its distribution along the line.

3) The recorversion of electrical into mechanical power, in the

car motor or locomotive.

The first of these, here in England at any rate, is dependent upon the economical production of steam power, although there are essential points of difference between the conditions under which steam-power is required for electric traction purposes and for electric lighting. But in Scotland and Ireland, and in many countries abroad, there is abundant water power, now only very partially utilised. The Portrush line is worked in part by water and in part by steam-power, but for the Bessbrook and Newry Tramway (of which there is a working model on the table) water-power is exclusively used.

A few experiments will show that the demand for power on the generating plant is greatest at the moment of starting the car or train, when, in addition to the power required to overcome the frictional resistances, power is also required to accelerate the velocity. Thus, if instead of a single car there are a number of trains moving on the one system, and it so happens that several are starting together, the demand made upon the generating plant may at one moment be three or four times as great as that made a few seconds after. This is shown in the diagrams which exhibit the variation of current supplied by the generators on the City and South London Railway, with eight trains running together, the readings being taken every ten seconds. The maxima rise as high as double the mean; thus the generating plant must be capable of instantly responding to a demand double or even treble the average demand upon it.

In electric lighting it is true there is not less variation between the maximum demand and the mean taken during the ordinary hours of lighting, but it is only in the event of sudden fog that the probable demand cannot be accurately gauged beforehand, and provided for by throwing more generators into action. Thus in a lighting station each generator may be kept working approximately at its full load, and therefore under conditions of maximum economy, whereas in a traction station the whole plant must be kept ready to instantaneously respond to the maximum demands which may be made upon it, and must therefore necessarily work with a low load factor, and consequently with diminished economy. So important is the influence on cost of production of the possible demand in relation to average demand, that the Corporation of Manchester, under their order for electric supply, have decided, upon the advice of their engineer, to annually charge a customer 31. per quarter for each unit per hour of maximum supply which he may require, in addition to 2d. for each unit actually consumed, i. e. for being ready to supply him with a certain amount of electrical power if required to do so, they charge an additional sum equivalent to the charge for its actual consumption for 1440 hours.

In one respect water-power has an economic advantage over steampower, because although steam engine and turbine alike work with greatly reduced efficiency at reduced loads, when the turbine gates are partially closed and the water restrained in the reservoir it is not