240 



SCIENCE. 



[Vol. VII., No. 162 



tween the diffracting edge and the observer's eye" 

 (Signal service, professional paper, xv. p. 41). 



So explicit a description as this from a well- 

 practised observer confirms the testimony of Euro- 

 pean specialists in sky colors, and leaves no question 

 whatever that Bishop's ring did. not then encircle the 

 sun. And yet, in the summer of 1884, it was so 

 strongly colored as to attract attention from the 

 guides in the Alps, and to call for special description 

 from more scientific mountain climbers. It was gen- 

 erally visible on clear days in the winter of 1884- 

 85, and on many favorable occasions through the 

 following summer. During this winter, it has seemed 

 to me to be generally less distinct than a year ago ; 

 but the most brilliant display that I ever recorded 

 was shortly after noon on the 2d of last November, 

 when the sun was hidden by a rather heavy sheet 

 of cirro-stratus cloud, while the western sky was clear. 

 The glaring and brassy central area was then en- 

 closed by a ring of strong reddish-gold color, fifteen 

 to twenty degrees from the sun ; next came the deli- 

 cate rosy or purplish pink, and at last the ordinary 

 blue of the sky. The colors were wonderfully vivid. 



Many if not most observers of the ring attribute it 

 to diffraction on particles of some sort derived from 

 the eruption of Krakatoa ; and, while this hypothesis 

 has much to recommend it, it cannot be denied that 

 the continued visibility of the ring puts a severe 

 strain on it. It is not to be wondered at that the 

 cosmic origin of the colors has its advocates, and 

 hence a method of determining the altitude at which 

 the diffracting particles float is of especial value. 



Dr. Zenker of Berlin has a pertinent article on the 

 question in a recent number of the Meteorologische 

 zeitschrift (Berlin, ii. 1885,400-400), in which he 

 discusses the effect that the altitude of the diffract- 

 ing layer of dust will have on the visibility of the 

 ring during total solar eclipses. And as a total solar 

 eclipse, visible in South America and on the Lesser 

 Antilles, will occur about half-past seven in the 

 morniug of the 29th of next August, we would re- 

 quest especial attention to this matter from astrono- 

 mers who may go down to observe it. Dr. Zenker 

 gives directions for observations on or near the cen- 

 tral line of the moon's shadow, and shows how they 

 may lead to the desired determination : for it is 

 evident, that, if the diffracting dust were all within 

 a few miles of the earth's surface, the colors of the 

 ring would fade away in a few seconds after the dis- 

 appearance of the sun ; while, if the dust lie far out- 

 side of the atmosphere, some portion of the ring 

 might remain visible during the whole eclipse. This 

 question will deserve a share of the watchfulness 

 generally given to the solar corona and infra-mercu- 

 rial planets. W. M. D. 



A trap-door spider at work. 



A trap door spider, Cteniza Californica, which 

 came from California in September, was put in a 

 box with earth, and soon made a nest with a perfect 

 door. She was found one morning occupying a hole 

 three-quarters of an inch in diameter and deep 

 enough to completely hide her, around which the 

 ground had been cleared and smoothed, so that it 

 was somewhat lower than the general level. Un- 

 fortunately, as this part of the work was done during 

 the night, she accomplished it unobserved. She 

 probably cleared the ground, however, as she had 

 done on a former occasion, when she was seen to 



walk slowly sideways, with all the feet on one side 

 held together, turning slightly at the same time, and 

 sweeping all rubbish and coarser bits of earth before 

 her. In digging the hole, she threw the earth to a 

 distance, as was shown by numerous little irregular 

 lumps of earth scattered over some moss at the 

 farther side of the box. Later the spider was seen 

 to dispose of more in the same manner, but it was 

 done so quickly that the exact motion could not be 

 distinguished. 



During the day she busied herself in the burrow, 

 apparently treading against the sides, in order to 

 make a compact wall. At night she rested, and 

 nothing more was done until the following evening, 

 when she commenced to build a straight ridge or rim 

 of earth at one side of the hole. She brought up as 

 much earth as could be carried under the mandibles, 

 and placed it on top of this rim. When it had been 

 secured by several strokes of the fangs, the spider 

 turned, and rubbed the spinnerets over the spot, and 

 afterwards all along the edge. The spinnerets were 

 applied directly to the surface, and were used not 

 only to produce the silk, but also to smooth and 

 model the edge. 



This process was repeated until the rim was about 

 a quarter of an inch in height, when the spider left 

 it, and commenced a similar one on the opposite edge 

 of the hole. Here she worked, as before, until she 

 had made a ridge about half as high as the other, 

 when she returned to the first, and during the next 

 hour added to them both alternately. At the end of 

 that time, she brought up the first load of earth 

 which was not used in building, and deposited it as 

 far away as she could reach, without leaving the 

 burrow. As she withdrew, she turned, and attached 

 a line of web to the edge of the second rim, by which 

 it was pulled over the opening after she had dis- 

 appeared from sight. Henceforth it was necessary 

 to lift and turn back this rim (or flap, as it might 

 now be called, to distinguish it from the true door) 

 whenever she came up, unless, as sometimes hap- 

 pened, she had neglected to pull it down. 



In the mean time, the first rim, which was to be- 

 come the true door, had been gradually enlarged ; 

 but another hour elapsed before any attempt was 

 made to pull it down. The spider then fastened a 

 line to the upper edge, by which, after a long and 

 steady pull from below, the structure was dragged 

 over the opening, which it only half covered. It was 

 immediately raised, and carefully re-adjusted in an 

 upright position. After another half-hour, devoted 

 to adding more earth to the two rims alternately, the 

 first was again drawn down; but, being still too 

 small, it was once more returned to the old position, 

 and the work of enlargement continued. As nothing 

 but persistence in this course seemed necessary to 

 complete the door, the spider was allowed to work 

 the rest of the night without supervision. 



In the morning the spider had vanished. The en- 

 trance of the nest was closed, and the depression 

 around it filled, so that its position was perfectly gen 

 cealed. Naturally, it was supposed that the door was 

 finished: but the next night proved this conclusion to 

 be erroneous. When the spider was visited at three 

 a.m., the door covered only three-quarters of the 

 opening, and she was still employed in adding earth 

 to the edge. During the day the entrance had evi- 

 dently been closed by the true door and the flap, used 

 together as a double or folding door, one side being 

 much larger than the other. The flap, no longer 



