848 TjiAysACTWxs of the Americax Institute. 



declination varies. But the mediaeval architects did not stop here. 

 They found out liow to so construct another line, a most difficult 

 curve, lying half east and half west of the straight one, that the sun's 

 image crosses this curve every da}^ at twelve o'clock noon. This 

 curve, resembling a very lengthened figure of S, is called the anal- 

 cniina. We may see it drawn now, on a small scale, but never well, 

 on the vacant part of the ocean of some terrestrial globes, its lower 

 or big loop standing on the south tropic, and its lesser loop reaching 

 the northern. If correctly drawn, the resemblance to an italic 8 

 would be so complete that it would lean a little from the meridian. 

 This figure then served (as it might again) all the purposes of our 

 " equation of time " tables. Few people know that the two maxi- 

 mum errors, whicli occur in jSTovember and February, when the clock 

 is furthest behind and before the sun, are not equal. One is nearly 

 seventeen minutes and the other is barely sixteen. Now, this arises 

 from the italic leaning of the 8, and is increasing and will increase 

 for many centuries to come. But about A. D. 1250 (from M^hich 

 date our globe-makers would seem to take their analemma unchanged, 

 though the loops were more unequal than now, the eight \vas upriglit. 

 Before 1250, it had leaned the otlier way ever since Adam's time. 

 But about his time it was not onl}' upright, but the loops, instead of 

 their being at their greatest inequality (as in 1250), were equal ; as 

 they became only once in about 13,000 years. Throughout these 

 18,000 years the southern loop continues the biggest ; and then, for 

 another 13,000 years the northern will be tlie biggest, and the present 

 changes will be repeated inversely. Thus the astronomer's tables, 

 that substitute the sun's longitude for montlis and days, although 

 called " practically" permanent, do not serve forever. 



In connection with this item on " sun dials," Mr. C. F. Boyle 

 remarked that the main difficulty Avith the correctness of the sun 

 dial is the difference of the refraction of the sun light at difterent 

 ])arts of the year. For instance, on a winter's day the light striking 

 obliquely there is more refraction, and hence the indications are 

 difierent in winter from what they are in summer. 



l\ i:ltot1':llus. 



Mr. Henry Wliitall exhibited a Ileliotellus, invented by Prof. J. 



Davis. Tliis instrument contains five globular bodies, which, by 



their arrangement, relations, and motions, exhibit the arrangement, 



relations and motions of the Sun, Mercury, Yenus, Earth, and the 



