194 



KNOWLEDGE 



[Sept. 5, 1884. 



moment that more pollen is beginning to exude from the 

 top of the anther tube, its exudation being due to the con- 

 traction of the sensitive filaments, which draw down the 

 anthers, and so cause the pollen to be expelled by the hairs 

 on the surface of the style. After continuing the irrita- 

 tation for twenty seconds or so, again brush off the pollen 

 from the top of the tube. You will find that the 

 anthers have now receded somewhat, and that the style, 

 with its branches still closely pressed together, begins to 

 [jrotrude slightly fi'om the summit of the anthers. In 

 other words, you have given the flower its cue for passing 

 from the first or male to the second or female stage. As 

 soon as that cue is once given, the change proceeds apace — 

 the anthers continue to shrink down into the corolla, and 

 the style to open its branches. 



The question of the mechanism for ensuring irritability 

 in the filament, I must leave to better microscopists than 

 myself. 



This is rather a more technical paper than usual, and I 

 fear some readers may find it difficult of comprehension. 

 But if they will take a sunflower in its prime, and cut it 

 in two, following the description with the living object, I 

 think it will all become quite clear to them. 



THE EARTH'S SHAPE AND MOTIO^^S. 



By Richard A Proctor. 



(Continued from page 176.) 



CHAPTER IV.— DETERMINING THE SHAPE OF THE 

 EARTH. 



AS our observer sets out on his northward journey he 

 sees the earth spread forth as a plane before him in 

 that direction as well as in others ; and, therefore, his 

 natural conclusion is that he is about to voyage in a 

 straight line. This, at least, is the assumption on which 

 all his observations must, in the first instance, be grounded. 



Let us suppose that an observer travels about 380 miles 

 towards the north before recommencing his observations of 

 the skies ; although not very important for the particular 

 observations now to be made, we shall yet suppose that he 

 has taken careful count of time, and has with him the 

 means of measuring time throughout his observations. 



He proceeds precisely as at his first station ; and, after 

 a similar series of observations, he finds results strikingly 

 similar to those before obtained. He finds the whole 

 heavens rotating in appearance about a fixed axis, with 

 perfect uniformity, and at precisely the same rate as he had 

 noticed when at his former station. But one important 

 difference attracts his attention. He finds that the point 

 about which the .stars appear to move (which point he may 

 determine either as before by watching the daily motion of 

 the sun in .spring or autumn — for we suppose him to settle 

 for some time at his new post — or by watching the motions 

 of the stars themselves) is higher up above the northern 

 horizon. Measuring carefully the extent of the change, he 

 finds that instead of being about 51^° the pole is now about 

 57" above the horizon. 



Let us see what this result seems to teach : — 



Let A (Fig. 1) be the first sta- 

 tion of the observer, B the second, 

 so that A B represents a distance 

 of about 380 miles. Then if the 

 assumption that the observer has 

 been travelling in a straight line 



is correct, he must draw AP at ' --. 



an angle of 5U° with AN, and 



B P at an angle" of 57°, these lines Fig. 1. 



of course meeting, and so forming a triangle A P B, whose 

 vertical angle A P B is the difference between 57° and 51^°, 

 that is 5i°. 



Now, before proceeding further, we notice here a source 

 of perplexity. 



When at A the observer found the whole of the heavens 

 moving exactly as though they were rotating around the 

 axis A P ; and now, when he is at B, he finds them moving 

 exactly as though they were rotating around the axis 

 BP. 



It need hardly be said that rotation cannot take place 

 simultaneously around two intersecting axes. Therefore, 

 we must select one of two alternatives. Either we must 

 abandon the notion of a uniform rotation, which seemed 

 in so simple and beautiful a manner to account for observed 

 appearances, or we must give up the theory that in travel- 

 ling along the earth surface the observer h;8 followed a 

 straight line. 



Now about the first of these alternatives, it is easy to 

 form an opinion. We know certainly that the line of 

 sight from A to a star traces out the surface of a right 

 cone, having A P as axis. We 

 know with equal certainty that the 

 same is true of the star as seen from 

 B. Hence, in place of the uniform 

 circular motion we deduced before, 

 we require that the star should travel 

 along the complex (and not even 

 plane) curve which is the intersection 

 of two right cones having intersecting 

 axes. But we need not follow out 

 the difficult considerations here suggested ; for we can take 

 the case where the cone becomes a plain, and so simplify 

 matters. 



Let E S W (Fig. 2) be the path of a star which rises in 

 the east, as seen from A. Then the line of sight from 

 A to this star lies always in the plane E A W S. But the 

 same star, as seen from B, also rises in the east and sets 

 in the west, attaining a lower altitude in the south 

 by 5°. 



Therefore its apparent path lies as e s w in the plane 

 e B ir s. This plane intersects the plane E A W S in a 

 straight line K L, and it is only by moving along the 

 straight line K L, starting with an infinite velocity from an 

 infinite distance in direction K, travelling with a con- 

 tinually diminishing velocity up to -, and thence passing 

 away with a continually increasing velocity to infinity to- 

 wards L, that the star could appear to move as it actually 

 does when viewed from A and B. The journey from 

 infinity in direction 2 K to infinity in direction 1, L would, 

 according to this view, occupy but half a day, and in some 

 inscrutable manner the star would return in the next half 

 day to its original position, at an infinite distance from us, 

 in direction 2 K. This is altogether absurd and in- 

 credible. 



Our observer, therefore, finds himself compelled to 

 abandon the theory that he has travelled in a straight 

 line from A to B. But, to make assurance doubly sure, he 

 applies other experiments. 



It will be seen at once (from Fig. 1), that if A B really 

 were a straight line, the distance of P could be at once 

 determined, either by a geometrical construction or by 

 calculation. Suppose the observer applies either method, 

 and then makes another journey equal in length to A B, 

 and still due northward ; then, renewing his ob.servations 

 of the heavens, he would have it in his power to make a 

 new calculation of the distance of P from A ; and if his 

 assumption that he was following a straight path were 

 correct, his rew estimate of A P should agree with his 



