i.^o 



NATURE 



[June i i, 1891 



shed were transferred from the South Ground to the 

 position formerly occupied by the old instrument, to 

 make way for the new buildings in the South Ground. 



The mean temperature of the year 1890 was 48°"6, 

 being o''6 below the average of the preceding 49 years. 

 The highest air temperature in the shade was 82°'8 on 

 August 5, and the lowest is"'! on March 4. This latter 

 is the lowest temperature registered in March since 1841, 

 being the same us that recorded on March 13, i?>45. The 

 mean monthly temperature in 1890 was below the average 

 in all months excepting January, March, May, and Sept- 

 ember. In December it was below the average by io^"o, 

 and in January above by ^"^'i. 



The mean daily motion of the air in 1890 was 272 

 miles, being 10 miles below the average of the preceding 

 23 years. The greatest daily motion was 837 miles on 

 January 26, and the least 32 miles on August 6. The 

 greatest pressure registered was 145 pounds on the 

 square foot on January 26. 



During the year 1890 Osier's anemometer showed an 

 excess of about three revolutions of the vane in the 

 positive direction N., E., S., W., N., excluding the turnings 

 which are evidently accidental. 



The number of hours of bright sunshine recorded 

 during 1890 by the Campbell- Stokes sunshine instrument 

 was 1255. which is about 35 hours below the average of 

 the preceding 13 years, after making allowance for 

 difference of the indications with the Campbell and 

 Campbell- Stokes instruments respectively. The aggre- 

 gate number of hours during which the sun was above 

 the horizon was 4454, so that the mean proportion of 

 sunshine for the year was o"282, constant sunshine being 

 represented by i. 



The rainfall in 1890 was 21 "9 inches, being 27 inches 

 below the average of the preceding 49 years. 



The winter of 1890-91 was remarkable for a long 

 period of exceptionally cold weather which commenced 

 on November 25, 1890. From this day till January 22 

 the mean temperature on every day except January 13 

 was below the average. The temperature was con- 

 tinuously below 32° on November 27, 28, December 10 

 to 19, 22, 23, 25, 28 to 30, January 2, 6 to 8, 10, 11, 17 to 

 19. The greatest defects from the average of 20 years 

 were on November 28 ( - I9°'i), December 22 (- 2o°7), 

 and January 10 (- I9"'3). The lowest temperatures re- 

 corded during the three months were i8°"3 on November 

 28, 1 3° -4 on December 22, and i2°-o on January 10. The 

 mean temperature of December 1890 was 29°"8, or lo^'o 

 below the average of the preceding 49 years, the coldest 

 December on record since 1841 previous to 1890 being 

 that of 1879, whose mean temperature was 32'^-4. In 

 this same month, December 1890, only 2'' 4 of sunshine 

 were recorded. 



Chronometers, Time Signals, and Longitude Opera- 

 tions. — The number of chronometers and deck watches 

 now being tested at the Observatory is 169 (113 box 

 chronometers, 20 pocket chronometers, and 36 deck 

 watches). The annual competitive trial of chronometers 

 commences on July 4, and the trial of deck watches on 

 October 24. 



The time-balls at Greenwich, Deal, and Devonport 

 are next referred to. 



The reductions for the longitude Paris-Greenwich are 

 now completed and ready for publication. In reference to 

 the discrepancy between the results of the French and 

 English observers, mentioned in the last Report, Com- 

 mandant Defforges visited Greenwich in June 1890, and 

 went carefully through the reductions with Mr. Turner and 

 Mr. Lewis. No mistake was found in the work, but 

 several questions of some importance were raised. The 

 results of the discussion and of subsequent correspond- 

 ence are summed up in two papers by Mr. Turner and 

 one by Colonel Bassot and Commandant Defforges, in 

 the Monthly Notices of the Royal Ast7'onomical Society, 



NO. I 128, VOL. 44] 



vol. li. pp. 155, 407, and 413 respectively. As the matter 

 now stands, the English definitive result for the difference 

 of longitude between the Greenwich transit-circle and 

 Cassini's meridian is 9'" 20^86, while the French result 

 (not yet published) is about o'*i5 greater, a discordance 

 which, though only about half of that found in the preli- 

 minary discussion, is still so large, that there seems to be 

 no alternative but to repeat the work with special precau- 

 tions suggested by the experience gained. 



The proposal to determine the longitude of Montreal 

 as the base station for the Geodetic Survey having been 

 sanctioned by the Admiralty last December, arrange- 

 ments have been made in concert with Prof. McLeod, of 

 the McGill College Observatory, Montreal, for a deter- 

 mination of the longitudes Montreal-Canso-Waterville- 

 Greenwich, the termini of the cable, Canso and Water- 

 ville, being occupied as longitude and not merely as 

 transmitting stations, a course which seems advisable in 

 view of the great geodetic importance of these points. 

 The necessary funds have been voted, and the Com- 

 mercial Cable Company have generously granted the use 

 of their cable. 



The determination of the longitude of Washington has 

 been deferred for the present. 



During the past year, Lieutenants Heming, Monro, 

 and Smyth, R.N., and Captain Haynes, R.E., have at 

 various times been instructed in transit-observing. Mr. 

 S. Hirayama, of the Tokio Observatory, was at work for 

 some weeks studying the general organization of the 

 Observatory. 



THE CLASSIFICATION OF THE TUNIC AT A 



IN RELATION TO EVOLUTION. 

 '"F'HE detailed classification of the Tunicata, and espe- 

 J- cially of the so-called " Compound Ascidians," has 

 usually been found a matter of special difficulty by sys- 

 tematists, and each successive investigator has discovered 

 grounds i^or modifying in important respects the grouping 

 of genera and families established by his predecessors. 

 A glance at the systems of Giard, Delia Valle, von 

 Drasche, and Lahille, all of recent date {i.e. post-Dar- 

 winian, and since the introduction of modern methods 

 and the recognition of the Tunicata as Chordata), shows 

 the notable want of agreement between competent au- 

 thorities. There is probably a special reason for this 

 exceptional diversity of opinion, and I believe the cause 

 is to be found in the course of evolution or phylogeny of 

 the group, and especially in the complex relations be- 

 tween the Compound forms and the other Tunicata. 



In fact, if the matter be regarded from the point of 

 view of the consistent evolutionist, the special difficulties 

 vanish, the complicated relationships between groups 

 (which can only be represented by dendritic diagrams, 

 or even in some cases by networks) become explicable 

 and natural, the great diversity in value of the as- 

 semblages of forms known as "genera" and " species" 

 is simply what would be expected, and the differences 

 between the various classificatory systems (allowing for a 

 few errors which have been corrected by later investiga- 

 tions) can be accounted for, and the conflicting opinions 

 of the authors reconciled. But, on the other hand, if the 

 subject be approached from the standpoint of the pure 

 systematist, whose object is to divide and subdivide into 

 clearly defined groups of approximately equal value, and 

 to recognize only "good" genera and species, nothing 

 but confusion results ; it becomes practically impossible to 

 distinguish and arrange naturally the groups of Simple 

 and Compound Ascidians ; and some of the most interest- 

 ing and instructive points, such as the gradation of 

 varieties into species and species into genera, and the 

 individual variations in specific characters, are altogether 

 lost sight of. 



These views were expressed partly in my Reports 



