May 14, 1874] 



NA TURE 



29 



about a second before the meeting of the cusps. The 



diminution of brightness is very sudden, and this is the 

 phenomenon to be chiefly attended to in the actual obser- 

 vation. It occurs almost exactly at the moment of true con- 

 tact, though the " black drop " docs not disappear until 

 some seconds later. It is of the utmost importance that the 

 nature of these different phenomena should be carefully 

 studied by all the observers. And at the present time 

 experiments are being made with a view of determining 

 the personal equation of each of the observers on the 

 British expeditions. 



But the actual observation will be rendered more diffi- 

 cult for various i-easons. Firstly, the enormous extent of 

 atmosphere which the rays of light must penetrate before 

 reaching the telescope will destroy the definition to a 

 large extent. Secondly, the existence of an atmosphere 

 around the planet Venus may materially affect the nature 

 of the phenomenon. 



In any case there is little doubt that as many of the 

 observers as possible of all countries should describe, as 

 accurately as can be done, the exact appearances which 

 are noticed at successive stages of the ingress and egress 

 respectively. Comparisons being also made between 

 different observers and between different telescopes, it 

 will be possible to reduce the observation of any phase 

 which may chance to be caught in the actual observation 

 to the true time of contact. From observations with the 

 Model Transit of Venus made at Greenwich, the follow- 

 ing facts appear : — 



1. It requires considerable experience for an observer 

 to appreciate all the definite changes of appearance which 

 occur. 



2. When two observers describe a particular phase 

 which they see, and determine to observe this phase to- 

 gether, the times recorded by each are generally accordant 

 within a fraction of a second. 



3. The successive phases of an ingress or egress ap- 

 pear to follow each other sometimes rapidly, at other 

 times gradually ; so that in some cases all the phenomena 

 are observed within three seconds, on other occasions the 

 same series of phases is completed in ten seconds. 



4. The time at which any particular phase is observed 

 varies very slightly with the aperture of the telescope. 

 When a telescope of good definition is employed, the 

 time of any phase at ingress is earlier than with an in- 

 strument of less perfect definition. 



In the case of the observations of last century, it is easy 

 to see how observers quite unprepared by previous ob- 

 servations as to the nature of the appearances they were 

 about to witness were sometimes inconsistent with each 

 other. In fact, without preliminary practice, and with bad 

 definition, observers might vary even with a Model 

 Transit of Venus by as much as 15 seconds. But, 

 knowing what they are to observe, they ^¥Ould differ 

 under no circumstances by more than about 2 seconds. 

 Hence it is probable that in the actual transit, if the de- 

 finition be good, the observation may be accurate to 

 within one second ; but if the circumstances be not very 

 favourable, they may differ to an extent of fully three 

 seconds, even after considerable practice with the model. 

 These estimates serve to give us some idea of the accu- 

 racy with which we may hope to h.ave the observations 

 made ; and it is probable, from the care which has been 

 taken to multiply the number of observers at each station, 

 that each pair of observations of contact will give us a 

 determination of the parallax of the sun true to about I 

 per cent. 



In the observations of contact, however, a great deal 

 depends upon the experience of the observer ; and it is 

 fortunate that the idea originally thrown out by M. 

 Janssen, and the mechanical execution of which has since 

 been so ably carried out, will indelibly record the pro- 

 gress of the phenomenon and serve as a check to the 

 observers. 



By the aid of this method photographs of particular 

 sun-spots have already been taken with great success at 

 intervals of one second during one minute of time. Each 

 of these sixty photographs is perfect in itself, and would 

 admit of very perfect measurements. Hence there is 

 every reason to believe that in this manner an indepen- 

 dent and very valuable observation of the true time of 

 contact will be made at each stationwhere a photo-helio- 

 graph is situated. 



The observations by means of photography during the 

 progress of the transit have few difficulties to contend 

 with. Their value will be largely increased by the fact 

 that the actual measurements will be made afterwards 

 when the observer cannot be carried away by the excite- 

 ment of the moment. But even in this class of observa- 

 tion there are difficulties which must be carefully con- 

 sidered. It is found that if a sensitised plate be over- 

 exposed, the image of the sun is considerably enlarged. 

 This is due to pliotogiapliic irradiation. It has been 

 found by Lord Lindsay and Mr. A. C. Ranyard to be 

 mainly due to the reflection of hght from the back of the 

 glass plate.* It can be almost entirely avoided by wet- 

 ting the back of the plate, and placing black paper against 

 it. There will still be probably a slight enlargement of the 

 sun's diameter. This will not affect the relative positions 

 of the centres of the sun and Venus ; but it will render it 

 extremely difficult to determine the unit of measure- 

 ment. 



There are two ways of applying the photographic 

 method. The first is the same as the heliometric method. 

 For this purpose it is necessary to have one station in the 

 north and another in the south. By the other method 

 we do not determine the least distance between the sun 

 and planet, but the actual position of the planet at each 

 observation. In other words, we determine the distance of 

 Venus's centre from the sun's centre, and also the angular 

 distance measured from the north point of the sun. To 

 do this we must have in the focus of the photo-heliograph 

 a fine thread to indicate the direction of the meridian in 

 the photograph ; or in the American method we must 

 have a thread suspended vertically which shall indicate 

 the vertical direction in the solar photograph. The 

 arrangements of the American method, as set up by Lord 

 Lindsayat Dunn Echt, are shownin Fig. 16. The siderostat, 

 lens, and hut, are all shown in position. . 



The value of the different methods has been well dis- 

 cussed by De la Rue,f Tennant.J and Proctor.§ The 

 method which takes into account tlie (/rf/z^/ position of the 

 planet on the sun is the more accurate, but it requires that the 

 fiducial lines, or lines of reference, shall be exactly repre- 

 sented in the photographs. Mr. De la Rue says that this 

 can be done to within one minute of space. 



Besides photographic irradiation, however, there is a 

 very important difficulty which enters into both the pho- 

 tographic and heliometric methods. This is due to the 

 refraction of our atmosphere. Everyone knows the dis- 

 torted forms which the sun assumes at the time of sunset. 

 In our own climate these appearances are seldom seen 

 on account of clouds and the haziness of the atniosphere. 

 But even from a high mountain, or from any position which 

 allows the form of the sun to be accurately seen up to 

 the time of sunset, its sh.ape may be noticed to be either 

 square, elliptical, or pear-shaped, accordmg to the cir- 

 cumstances of the atmosphere. Now at the most favour- 

 able points of observation the sun will be comparatively 

 near to the horizon. Consequently its form will vary with 

 the temperature of the air and with atmospheric disturb- 

 ances. With our feeble knowledge of the laws of refrac- 

 tion it will be a matter of some difficulty to dttermine 

 with accuracy the distance at different times between the 

 centres of the sun and Venus. 



« Monthly Noticts of the R. A. S 1873, p. 313. 



t Ibid, xxix., 48 and =82. J Il'id. 280. 



§ IHd. XXX., 6s. 



