May 28, 1 891] 



NATURE 



colours. Stenopus hispidus, though so very conspicuous when 

 taken out of the water, was extremely difficult to see in the pool 

 where I found it. I should, in all probability, have failed to 

 notice it, had I not quite unintentionally and blindly touched it 

 with my stick. Like all animals protected by concealment 

 colours, it remained perfectly motionless when alarmed. When 

 looking down on to the growing edge of a reef from a boat on a 

 calm day, it is very difficult at first to see anything but the 

 corals and sponges. After a time, when the eyes become more 

 accustomed to the light, the fish may be distinguished. Those 

 that are coloured blu2 are much less readily seen than the gold, 

 yellow, and red varieties ; but an examination of the fish that I 

 caught myself, and were caught for me by the natives, showed 

 that the fish in which blue is the prevailing colour are much 

 more frequent in the very shallow water, while those that were 

 caught in water from 15 to 20 fathoms were more frequently red 

 or yellow. The blue colour seems to be a protection for the 

 fish from air-breathing enemies — the eagles, ospreys, and hawks 

 • — and as these enemies can only approach them from above, the 

 colours are frequently confined to the dorsal sides. The red and 

 yellow colours of the fish seem to be a protection from animals, 

 such as the sharks, perch, and other carnivorous fish, that 

 approach them from the deeper waters beyond the reefs. Thus 

 red and yellow fishes rarely have these colours confined to the 

 upper sides, and many of the blue fishes are coloured red or 

 yellow ventrally. 



It is difficult to frame any general rule to account for the 

 curious distribution of the colours of these animals in spots and 

 stripes. Speaking in very general terms, for there are many 

 exceptions, the fish that browse on the corals, possessing small 

 mouths and chisel-shaped teeth (such as the Chretodons, Trigger 

 fish, and Surgeons), are striped ; those that feed on other fish, 

 and have large mouths armed with^carnivorous teeth, such as 

 the Serranidse, are spotted. 



The only example of what appears to be a warning colour that 

 I have noticed occurs in connection with the spines on the tails 

 of certain Surgeons and Trigger fish. Acanthurus achillis, for 

 example, has a uniform purple colour, but there is a bright red 

 patch surrounding the formidable tail spines that give these fish 

 the name of Surgeons. Similar warning colours are very pro- 

 nounced also in Naseus unicornis and Naseus lituratus, and 

 in some of the Balistidse. 



WASHINGTON MAGNETIC OBSERVATIONS, 

 1886.1 



'T'HIS volume contains the results that have been obtained 

 from the magnetic observations taken at the Naval Obser- 

 vatory during the years 1888 and 1889. The instruments with 

 which they were made were, in the year 1887, placed in their 

 respective buildings that had been erected for that purpose by 

 the Bureau of Navigation. In the construction of these build- 

 ings the greatest care was taken to insure the complete elimina- 

 tion of local disturbances. No iron or any magnetic material 

 was used at all, and the fastenings, &c., were entirely of copper, 

 brass, and wood ; even the stoves, in which only wood was 

 burnt, were of soap-stone, with copper pipes. 



The instruments that were employed consisted of a declino- 

 meter, theodolite, portable magnetometer, dip circle, a set of 

 self-recording magnetographs, a seismoscope, and seismograph ; 

 each of them, with the exception of the last two mentioned, 

 being set on piers based on concrete, and in no way connected 

 with the floors of the buildings. To complete the equipment, a 

 compass-testing stand is placed on a pier north of the theodolite, 

 and is capable of motion in an east and west direction. By 

 means of an arm carrying two prisms that have adjusting screws, 

 the opposite marks on the compass card can be placed in the 

 field of view of the theodolite when the latter is directed on the 

 prisms. All the observations, which are represented in tabular 

 form, denote the results that have been obtained after applying 

 all necessary corrections. The tables include, among others, 

 the mean hourly values of the horizontal and vertical force for 

 each month of 1889, and of the declination for each month of 

 1888 and 1889, the last of which are taken from the monthly 

 curves ; declination ordinates for each hour, in minutes of arc 

 taken from daily declination traces ; hourly values of horizontal 



' Appendix I.—" Magnetic Observations." By Ensign J. A. Hoogewerff, 

 U.S. Navy. (Washington : Government Printing Office, 1890.) 



NO. I I 26, VOL. 44] 



and vertical force in absolute measure with all corrections ; ob- 

 servations of horizontal intensity and dip, with a summary of 

 disturbances in declination which differed two minutes or more 

 from the mean monthly curve. 



No less important is the series of the fourteen large plates at 

 the end of the volume. The first shows the way that the daily 

 photographic traces of declination, horizontal and vertical force 

 are recorded ; while the second illustrates the mean diurnal 

 variation of the magnetic elements for the year 1889. In this 

 latter plate the curve that gives the integration of these elements 

 — that is, that gives the mean diurnal total force — brings out the 

 fact that in every twenty-four hours there are two maxima and 

 two minima, these latter two occurring between midnight and 

 noon (75th meridian mean time). 



Plates iii. to vi. inclusive show the traces of the monthly 

 composite curves of declination for the two years. 



In Plates vii. to xiv. most interesting comparison is made of 

 the disturbed days of declination taken from observations at 

 Washington, Los Angeles (California), Toronto (Canada), and 

 Pawlowsk (Russia) : the curves are all computed for the same 

 time (i.e. for the 75th meridian west of Greenwich), and reduced 

 to the same length of base line. Although on the whole the 

 curves show a more or less equal variation, yet there are some 

 cases in which a decided local variation has taken place. For 

 instance, on January 20, between the hours of noon and four 

 o'clock (75th meridian time), the magnetic declination at Wash- 

 ington, Los Angeles, and Toronto, shows only slight variations, 

 while at Pawlowsk the disturbance is in comparison quite large. 

 Another very interesting case happens on March 17, when the 

 curves traced at Washington and Toronto are quite similar to 

 each other, but different from those traced at the other two places : 

 the curve showing the magnetic disturbances in declination at 

 Pawlowsk being very similar to that indicating the horizontal 

 force at Washington. 



UNIVERSITY AND EDUCA TIONAL 

 INTELLIGENCE. 



Cambridge.— The subject of the Rede Lecture, to be given 

 by Sir Alfred Lyall on June 17, is "Natural Religion in 

 India." 



The General Board of Studies have again brought forward 

 proposals for the increase of the stipends paid to University 

 Lecturers and Demonstrators in Natural Science, which had to 

 be postponed last year owing to want of funds. < 



Mr. A. Hutchinson, Demonstrator of Chemistry in Caius 

 College, has been recognized as a Teacher of Chemistry with 

 reference to the regulations for medical degrees. 



A Syndicate is proposed by the Council of the Senate for the 

 purpose of considering whether any alternative for Greek should 

 be permitted in the Previous Examination. This is sure to 

 rouse much agitation, but it may be hoped that the long-vexed 

 question will at length be settled in a liberal sense. 



Another Syndicate is to consider the office of Superintendent 

 of the Museums of Zoology and Comparative Anatomy, about 

 to be vacated by Mr. J. W. Clark, Registrary. Some rearrange- 

 ment of the duties, &c., is considered desirable. 



SCIENTIFIC SERIALS. 



American journal of Science, May. — On the relationship of 

 the Pleistocene to the pre-Pleistocene formations of the Missis- 

 sippi basin, south of the limit of glaciation, by T. C. Cham- 

 berlain and R, D. Salisbury. — On certain measures of the 

 intensity of solar radiation, by William Ferrel. The author 

 shows that many measures of the intensity of solar radiation are 

 of uncertain value. He specially discusses M. Crova's curves of 

 the relative intensities of solar radiation, obtained at Montpellier 

 with a modified form of the thermopile, called the registering 

 actinometer. — Geological age of the Saganaga syenite, by 

 Horace V. Winchell. — On a self- feeding Sprengel pump, by 

 H. L. Wells. — Contributions to mineralogy. No. 50, by F. A. 

 Genth ; with crystallographic notes by S. L. Penfield and L. V. 

 Pirsson. The composition and habits of the following minerals 

 are given : three new varieties of axinite, eudialyte, and 

 monticellite, and titanite from Magnet Cove, Arkansas. — 



