GO 
MAGAZINE OF SCIENCE AND ART. 
The water is rendered opaque by a weak solution of 
sulphate of indigo, and is introduced in sufficient quan¬ 
tity to fill about two-thirds of the bulb. The instru¬ 
ment being then cleared of air by continuous boiling, is 
completed by hermetically sealing the extremity of the 
graduated tube. 
To use this sun-gauge, the whole of the liquid must 
first be poured into the bulb, which can be readily done 
on account of the bent position of the interior part of 
the tube. It is then inverted and placed in some sup¬ 
port where the sun’s rays may fall upon the bulb in¬ 
terruptedly throughout the day, while the tube beneath 
is completely shaded from the sun, but freely exposed to 
a current of air. 
It is evident that, as long as the sun is above the 
horizon, and the heat of the rays is at all preemptible, 
water will evaporate from the bulb and condense 
colourless and pure in the graduated tube, by the 
divisions of which its amount may be quickly and 
easily read off. The quantities of water distilled over 
will be roughly proportional to the total amounts of 
heat falling upon the bulb. 
This rough model is subject, however, to many 
serious defects, the chief of which is that, when evapora¬ 
tion has proceeded to a considerable extent, the quan¬ 
tity of liquid remaining in the bulb, is of course 
diminished, while the tube is partially filled. Thus 
both the absorbent and condensing surfaces are lessened 
in comparison with the evaporating surface, and the 
distillation of a millemetre’s depth of water will in¬ 
dicate less amount of heat rays at the commencement of 
the exposure than towards the end of the day. 
A more complicated form of the instrument (see Jig. 
II.J is inrended to remedy these and other defects, 
hut as tliis instrument exists as yet only on paper, it is 
a matter of doubt whether practical difficulties would 
not he encountered in its construction. 
It consists of three distinct parts, viz. 
1 . The evaporator. —A. 2. The condenser. —B. 
3. The register tube ,—C, 
The evaporator is, of course, exposed to the sun, but so 
that no rays fall upon any part of the included liquid 
but the upper surface, which for that purpose is placed 
exactly on a level with the suirounding screen. The 
condenser is connected with the evaporator by a wide 
bent tube, something in the manner of a retort, but is 
placed several indies above the evaporator. The lower 
parts of both these bulbs are also connected by tho 
register tube, which is of narrow' bore (about one-tenth 
of an inch) and of considerable, in fact, of any desirable 
length. Being open at both ends it serves to conduct 
any water collected in the condenser back again into the 
evaporator from which it originally passed in the state of 
vapour. Yet the amount of evaporation is clearly marked, 
as it proceeds, by a small globule of mercury (D) which 
is placed in the narrow tube, and as it does not allow' 
any water to pass, is forced along with tho column ; 
if then, tho index globule of mercury he at the end of 
tho register tube next to the condenser, when the instru¬ 
ment is first exposed to the sun, the length of the 
column of distilled water between it and the condenser 
will iudicate at any subsequent moment the amount of 
evaporation, or, in short, of radient heat absorbed by the 
evaporator. 
It is evident that as the water runs freely hack from 
tho condenser into the evaporator, the .surface ofwater in 
the latter w ill remain invariable, or very nearly so—not 
quite so, indeed, for as mercury is fourteen times as 
heavy as water, the index globule will balance a column 
of water fourteen times its own length, so tint the ex¬ 
tremity of the column of water in the register tube will 
be caused to oscillate about \\ inch above or below the 
level of the water in the evaporator according as the in¬ 
dex globule is in an ascending pr. descending bed of the 
register tube; this, however, will scarcely affect the 
level of the water in the evaporator, and the length of 
the column of distilled water may be easily ascertained 
by reading the position of both its extremities on the 
divided scale and taking the difference. 
The instrument is to be set for observation by first 
pouring the mercury and water into the condenser, then 
separating off the water through the wide tube into the 
evaporator, when upon setting it again in the usual 
position, the index will take its place at the head of the 
register tube. The glass-work ts fixed, (as shewn in 
Jig. III. J on a frame of wood, covered by a curved non¬ 
conducting screen, (marked h\) and placed for use in a 
case made of three sides only, which we may term the 
exposing case. 
It is clear that of the total amount of heat-rays fall¬ 
ing upon the bulb of a san-gauge, only a part will reach 
and be absorbed by the evaporating surface, the remain¬ 
der being reflected or absorbed by the glass. This is a 
defect, perhaps, unavoidable in any instrument applied 
to measure radient heat, upon w hatever principle con¬ 
structed. In the same individual sun-gauge, however, 
the results will probably bo proportional to the total 
amounts of radiant heat’ falling from the sun, and dif¬ 
ferent sun-gauges being compared together or with a 
standard instrument in actual use, correction factors 
may be determined which will reduce all results to 
complete uniformity. This experimental comparison is 
necessary with Sir J♦ llerschel’s aetinometcr. 
The absorbent surface of water being perfectly hori¬ 
zontal, intercepts an extent of the sun’s rays, which will 
be proportional to the sine of the angle of elevation of 
the sun. Thus the full intensity of the sun’s heat will 
be indicated only when the rays fall vertically; and 
when the sun is on the horizon, no ravs at all will tall on 
a horizontal surface. This, however, is rather an advan¬ 
tage than otherwise in meteorology, since the precise 
quantities of heat are thus represented which tall in like 
manner on the horizontal surface of the ground. 
It is very difficult with the sun-gange, as with any 
other sort of actinometer, to prevent disturbance from 
heat reflected or radiated from neighbouring objects, 
such as parts of the case which is necessary to hold and 
shelter the instrument. This is an objection still to be 
remedied, . . 
Amounts of radiated solar heat may he conveniently 
expressed by the depth (in parts of an inch or in miHe¬ 
rn etres) of evaporation that they will produce troin a 
surface of water exposed with periect freedom to his rays, 
under the condition that all aqueous vapour ol a greater 
tension than *109 inches of mercury (the tension at 32 
degrees Fnhr.) shall be instantly conveyed away. Or 
the numerical ratio of this mitt inch of water to tho 
aotine &or the “abstract unit of solar radiation ” adopted 
by Sir J. Herschel,having been once determined, all the 
results could of course be equally well expressed ac¬ 
cording to this unit. In practice, tho vapour in the 
sun-gauge will not be conveyed over into the condenser 
or sheltered part of the apparatus till its tension rises 
above that due to the temperature of that bulb, which is 
tho same as the temperature of the air. As, however, 
the latent heat of vapour diminishes as the sensible heat 
rises, a small correction must ho made according to the 
mean temperature of the air during the time of obser¬ 
vation, as given in Table I. of the Appendix. 
A thermo meter, T, is inserted in tho evaporator ot 
Figure IT., with its bulb just beneath the evaporating 
surface. The readings of this thermometer should agreo 
almost exactly with those of the temperature of the sur¬ 
rounding air, since all the sun’s heat, as I have beiore 
explained, is instantly conveyed away by evaporation. 
It is evident that as we increase or dimmish the area 
of the surface of evaporation, the weights of water caused 
to evaporate by the same intensity oi radiantheat, will 
increase or diminish in the same proportion, bupposmg, 
then, the capacity of tho divisions of the register-tube to 
remain always the same, we must construct the instru¬ 
ment with a Very larire evaporating bulb, it we wish it 
to bo very sensitive^ or to giv e large indications to a 
* The actine is “that intensity of solar radiation 
which at a vertical incidence, and supposing it wholly 
absorbed, would suffice to melt cnt-millionth part of a 
metre in thickness from the surface of a sheet of ice 
horizontal^ exposed to its action per minute of mean 
solar time .’’—Admiralty Manual , page 30,. 
