212 



NATURE 



[June 30, 1898 



Nicol's prism, thence through a long tube with plane glass ends 

 containing liquid carbon bisulphide, and afterwards through a 

 second Nicol's prism. The lay of light is received on a screen 

 having a sensitised surface, which is carried forward at an 

 uniform speed ; a long coil is wound round the tube containing 

 the carbon bisulphide, the prisms being adjusted so that no light 

 passes through the tube when no current is flowing through the 

 coil, the source of light being an arc lamp. 



The passage of a current rotates the polarised ray within the 

 tube, and the light then falls on the sensitised screen, and is 

 thereby recorded. 



As neither of these methods of reception are suitable for every- 

 day use, the British Post Office undertook, in conjunction with 

 the inventors, a series of valuable and interesting experiments 

 over the departmental lines under more practical conditions. 

 The existing departmental records of capacity, resistance and 

 mileage, compiled for the whole country, proved invaluable by 

 supplying exact data for each of the experiments performed, and 

 enabled reliable tables and curves to be constructed. The 

 experiments consisted of determinations of the highest limits of 

 speed for the Wheatstone automatic, as well as the synchrono- 

 graph system on various lines, the following combinations being 

 specially compared : — 



(i) Ordinary Wheatstone automatic with condenser compensa- 

 tion as is used at present. 



(2) The synch ronograph sine wave transmission system with 

 chemical receivers. 



(3) A combination of the synch ronograph sine wave trans- 

 mission with Wheatstone receivers. 



The alternator used for these experiments consisted practically 

 of several separate alternators on one shaft, each being inde- 

 pendent of the remainder, and so constructed that, with the same 

 speed of revolution, different frequencies or wave speeds could 

 be obtained, transformers being used in those cases where it was 

 desirable to maintain the E. M.F\ unaltered. 



Careful estimations were made not only of the force employed, 

 which is about 50 per cent, higher than that ordinarily used on 

 Wheatstone circuits, but also of the wave speed, and its 

 equivalent value in " words per minute" in each case. 



On a line from London to York and back, mainly composed 

 of copper, having a total mileage of 431!, and a K.R. equal to 

 33,000, a speed of 540 words per minute was attained with 

 Wheatstone receiver and Crehore-Squier transmitter (synchrono- 

 graph), although the maximum limit was not reached. The 

 speed obtainable with this K. R. being only 360 when ordinary 

 Wheatstone automatic was used. 



From London to Aberdeen and back, with a total mileage of 

 I097i and a K.R. of 261,000, a speed of 135 words per minute 

 was obtained by the Crehore-Squier Wheatstone combination, as 

 compared with 46 words per minute on the ordinary Wheatstone 

 automatic with the best compensating arrangements. 



These two cases are typical of the whole series of observations, 

 which enabled the comparative wave speeds of the different 

 systems to be estimated as follows : — 



W^heatstone automatic alone ... ... ... ... i 



Crehore-Squier transmission and Wheatstone receiver ... 2-9 

 Crehore-Squier transmission and chemical receiving ... 2 "9 



In the first two cases the number of waves necessary for each 

 word is of course the same, but in the last-named case, where 

 chemical receiving is employed, a further gain is obtained by 

 using fewer waves for each word, making the word speed in the 

 three cases bear the ratio i, 2-9 and 7. 



Chemical receiving is by no means so convenient as ordinary 

 Wheatstone, and the most pressing practical requirement at the 

 present day is not higher speeds for short distances, but higher 

 direct working speeds over long lines where at present inter- 

 mediate "repeaters" are necessary. 



It is satisfactory to note that the maximum wave speed 

 attainable by synch ronograph transmission with the chemical 

 receiver or with the Wheatstone receiver is exactly the same on 

 any circuit where the speed is limited by the line itself and not 

 by the receiving apparatus. 



On the Wheatstone system shunted condensers are necessary to 

 compensate for two distinct effects — the unequal duration of the 

 signals, and the inductance of the receiver. Where the syn- 

 chronograph transmission is employed on short cables or open 

 lines, no line compensation is required, and a fixed condenser can 

 be shunted across the receiver coils so as to compensate for the 

 inductance of the receiver for any given speed. In connection 



NO 1496, VOL. 58] 



with this question the inductance of the Post Office receiver was 

 carefully verified, and was found to be 3 •46 Henrys, the necessary 

 condenser compensation depending solely on the speed of trans- 

 mission (or wave-frequency) and the arrangement of the receiver 

 coils, and in no instance having any direct or complicated relation 

 to the line capacity. 



On an artificial cable, equal to about 200 miles of ordinary 

 submarine cable, where condenser compensation is used at both 

 ends, the increase of wave speed obtained by the synchronograph 

 was imly 50 per cent, instead of 190 pec cent, as in the case of 

 open wires. It would therefore appear that with further ex- 

 periment some line compensation might be found to be necessary 

 for cable working. 



The experiments show that where the capacity of the line is 

 not great, as in the case of aerial lines, the transmission 

 of the current in sine waves produces the best results, and 

 leaves the factor of the inductance of the receiving instrument 

 to be dealt with separately, and consequently in a more exact 

 manner. 



The principal difficulty in the application of the system 

 is the necessity for the use of a new code of signals, or a 

 reduction in the speed value to admit of conformity with 

 existing codes. The existing Wheatstone automatic instruments 

 are also light, portable, and adapted for use in outlying districts 

 at short notice, where the synchronograph would probably be 

 found to be less suitable. The perforator at present in use 

 for the preparation of the transmitting "slip" has also, by 

 a process of evolution, become extremely convenient, and 

 equally suitable for hand working in confined spaces, or where 

 power is available. 



A suitable and easily manipulated perforator for the syn- 

 chronograph has yet to be devised. Messrs. Squierand Crehore, 

 however, deserve great credit for the discovery, with limited 

 means of experiment, of an improved and promising system of 

 high speed transmission. 



OBSERVATIONS ON S TOM ATA} 



T^HE method described depends on the fact that in adult 

 -'- leaves transpiration is stomatal rather than cuticular, so 

 that, other things being equal, the yield of watery vapour 

 depends on the degree to which the stomata are open, and may 

 be used as an index of their condition. In principle, it is the 

 same as the methods of Merget {Comptes rendus, 1878) and 

 Stahl (Bot. Zeitiing, 1894). These observers used hygroscopic 

 papers impregnated with reagents which change colour according 

 as they are dry or damp, and Stahl, who employed paper soaked 

 in cobalt chloride, has obtained excellent results. In my labora- 

 tory I have used, for some years, a hygroscope for demonstrating 

 ■cuticular transpiration, in which evaporation is indicated by the 

 untwisting of the awn of Stipa pennata (Darwin and Acton, 

 " Practical Physiology of Plants," ist edition, 1894) ; my present 

 instrument is of the same general type, but the index is made of 

 " Chinese leaf," i.e. shavings of pressed and heated horn.^ 

 If a strip of horn is placed on a dry substance, e.g. the 

 astomatal surface of a leaf, it does not move, but on the stomatal 

 surface it instantly curves strongly away from the transpiring 

 surface. In the hygroscope the degree of curvature is read off 

 on a graduated quadrant, and in this way a numerical indication 

 of the condition of the stomata is obtained. 



The instrument makes no claim to accuracy, but has proved 

 extremely useful when used comparatively to indicate and localise 

 small changes in the transpiration of leaves, and therefore, by 

 implication, changes in the condition of the stomata. By 

 observing under the microscope the uninjured leaf of Caltha 

 pahislris, and comparing the variations in the size of the 

 stomata with the variations in the readings of the hygroscope, 

 it is easy to convince one's self of the value of the method. It 

 must be especially noted that though a fall in the hygroscope 

 readings corresponds with a narrowing of the stomatal opening, 

 it does not follow that zero on the hygroscopic scale means 

 absolute closure of the stomata. This want of sensitiveness 

 has one advantage, namely, that cuticular transpiration has no 

 effect on the horn index, so that any movement of the index 



1 A paper by Francis Darwin, F.R.S., read at the Royal Society, 

 June 16. 



2 I also use the epidermis of a Yucca — a material which I owe to the 

 kindness of Mr. Thiselton-Dyer. 



