DD) 
=, 
Fuly 1886 | 
for the benefit of travellers, in which the floating glass 
was part of the plan. 
In obtaining the position of a ship at sea the difficulty 
is to get observations both for latitude and longitude at 
the same time, as all other observations depend on the 
distance covered by the vessel in the time which has elapsed 
between the observations. Now, as this distance depends 
not only on the direction and rate of the vessel through 
the water, but also on the direction and rate at which the 
water itself is moving, and as this latter element in the 
calculations cannot be ascertained with precision, it 
follows that all observations at sea which depend on the 
ship’s run in the interval have an element of uncertainty. 
The best time to obtain simultaneous observations for 
latitude and longitude is at twilight, morning and evening, 
as then the horizon is clear, and, unless the weather is 
very cloudy, some stars can be seen. Here Sumner’s 
method is invaluable, as three or more stars can be 
utilised and the correctness of the result guaranteed, pro- 
vided, of course, that the chronometer is correct. In the 
day-time the only chance to obtain simultaneous observa- 
tions is when the sun and moon are both visible, or when 
Jupiter, or Venus, happen to pass the meridian at an 
interval of over 23 hours from noon, as then, in bright 
weather, their meridian altitudes can be obtained by a 
practised observer with a good sextant. 
One of the difficulties in obtaining good results at sea 
is owing to the varying nature of the refraction, more 
especially close to the horizon. This may be guarded 
against in the case of the meridian altitude of the sun by 
observing, when practicable, its altitude with the north 
and south horizons. To show the closeness of the results 
ascertained in this manner, it is only necessary to observe 
that H.M.S. 77z¢on, when fixing the position of the Ower 
and Lemon light-vessel on the east coast of England in 
1884, obtained the latitude on four different days, the 
results being as follows :— 
ia 25 Lat. 53 z 56 X. 
uly 9 on 33 oN. 
July 1 » 537 54N. 
July 12 pp ish 7 BYP IN 
an extreme range of 6”, or 600 feet, in the latitude. Such 
a close accordance shows the value of this method, which 
is recommended by Raper. 
As regards obtaining the longitude by lunar distances, 
this has been gradually falling into desuetude owing to 
the quicker passages made by vessels and to the cheap- 
ness of chronometers. There can, however, be no doubt 
of its utility, as it is the only good way of obtaining the 
position of the ship at sea should any accident happen to 
the chronometers, and it is to be regretted that it is so 
seldom practised, particularly when we remember the 
excellent results obtained by the older navigators, es- 
pecially by Cook. For the actual observation the 
repeating circle is a far better instrument than the sex- 
tant, as by it the distance between the sun and moon is 
observed with much greater accuracy, a matter of the 
utmost importance when we remember that an error of 
_ one minute in the distance makes an error of twenty-five 
miles of longitude under the most favourable circum- 
stances. It is therefore evident that this observation 
requires to be made with the utmost care and that con- 
stant practice is necessary to obtain good results. 
. 
NATURE 
263 
In the problem of obtaining the true bearing of a ter- 
restrial object from a ship at sea, Mr. Merrifield has 
omitted the correction of the angular distance due to the 
height of the object: this is probably an accidental omis- 
sion, but although it does not usually amount to much, it 
is desirable the student should be acquainted with it. 
LETTERS TO 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 ts taken of anonymous communications. 
[The Editor urgently requests correspondents to keep their letters 
as short as possible. The pressure on his space ts so great 
that it ts impossible otherwise to insure the appearance even 
of communications containing interesting and novel facts.] 
Laws of Solution 
IN my paper on ‘‘ Chemical Affinity and Solution,” published 
in NATURE, vol. xxxili. p. 615, I gave some general proofs 
(taken from Thomsen’s researches on thermo-chemistry) of the 
truth of my theory of solution. I shall now show that there are 
| certain well-marked and definite laws of solution which are in 
complete accord with that theory, and seem to me to place it 
beyond doubt. In all chlorides, bromides, iodides, sulphates, 
and nitrates, for which data are available, the heats of solution 
in water vary directly— 
(1) As the affinity (measured by heat of combination) of the 
positive element of the salt for O varies ; 
(2) As the affinity (measured by heat of combination) of the 
negative element or radicle of the salt for H varies ; 
And znversely— 
As the affinity (measured as above) between the positive and 
negative elements of the salt varies. 
The following examples will make this plain :— 
Heat of : Heat of 
Compound combination Difference solution of Difference 
Bs chloride 
[Mg,Cl, I51010 = 35920 — 
[Mg, 0, Aq] 148960 2350 —- — 
[Ca,Cl,] 169820 — 17410 + 18510 
[Ca,O, Aq] 149260 20560 — — 
— 18510 — +18510 
[Ca,Cl,] 169820 — 17410 a 
[Ca,O, Aq] 149260 20560 — — 
[Sr, Cl, ] 184550 — 11140 +6270 
[Sr,O, Aq] 157780 26770 — 
— 6210 — +6270 
[SrCl,] 184550 — II140 = 
[Sr,0,Aq] 157780 26770 — — 
[Ba,Cl,] 194740 _— 2070 +9070 
[Ba,O, Aq] 158760 35980 — -- 
-— 9210 — +9070 
Similar results are obtained if we substitute the alkali metals 
for above, but there is a variation in the case of metals which 
form insoluble oxides or hydrates. In the latter case the heats 
of solution are not so great as they should be if compared with 
above compounds. Among themselves, however, they follow 
the laws pretty closely, and seem arranged in groups. Thus, 
ZnCl, and CdCl,, FeCl,, CoCl,, and NiCl, form two such 
groups. 
The foregoing examples illustrate the effect of the change of 
the positive element of the salt on the heat of solution. Now 
let us change the negative element and we shall see the same 
result. 
