of the Fishery Board for Scotland. 
swept north oiid south by the tides, and incorporated with the mas8 of 
sea water. In ordinary circimistancos the Bell Eock, although almost 
directly opposite the nioutli of the Tay, and distant from it only 8 
miles, a[)pears to be quite uualfected by the water discharged. This 
suggests the possibility of a salinity limit being traced at a certain dis- 
tance from shore, which might possibly separate one species or age of 
pelagic organisms from an'.)ther and by affecting, the distribution of fishes' 
food, tluis affect the distribution and movement of the lishes also. 
Abertay Light-vessel. — This vessel is anchored in about 8 fathoms 
of water, the depth varying from 6 to 10 fathoms according to the 
tide, at tlie mouth of the Firth of Tay. Its position differs from that 
of any of the other stations, most nearly resembling Oxcar, but subject 
to much more intense and variable conditions. The tides run very 
strongly, and the shifting of the sandbanks on both sides of the entrance 
to the Firth must produce corresponding changes in the direction of 
the tidal streams. Observations were commenced on June 5, 1889, 
and as the observers had been previously trained to similar work, the 
series is not only the longest but tlie most trustworthy wliich the 
Board has secured. It is also remarkably continuous, the interruptions 
being far apart and of short duration. Temperature and specific gravity 
were observed for the first few months at 10 a.m. and 4 p.m., latterly 
at 9 A.M. and 3 p.m. as at other stations, and they were taken at 
surface, 1 fm., 3 fms,, 5 fms., and bottom. The entire series of 
readings have been discussed in exceptional detail, and have yielded 
many results of interest. 
The mean surface temperature from June to December 1889 was 
10°'9 at 10 A.M., and 11° -1 at 4 p.m. ; the mean bottom temperature 
was 10° -8 and 10° '9 at the respective hours. The mean surface 
temperature for 1890 was 8°-9 at 9 a.m., and 9°'0 at 3 p.m. ; the 
bottom means at the same hours being 8° "9 and 8° '9. For the first 
three months of 1891, the surface temperature averaged 4° '5 at 9 a.m., 
and 5°'0 at 3 p.m. ; the bottom temperature at the same times being 
4°*7 and 4°*9. This shows an average rise of 0°*2 from 9 a.m. to 3 
P.M. on the surface, and rather less than 0°*1 at the bottom; and on 
the average the temperature of surface and bottom water is almost 
exactly the same. Looked at in the lump, then, the Abertay observations 
show very little effect of the sun in raising the temperature of water 
through the day, and very little difference at various depths. 
The highest surface temperature recorded was 14°'9 at 9 a.m. on 
August 10, 1889, the next 14°*5 at 3 p.m. on August 8, 1890. The 
lowest surface temperature was 2°'4 on January 10, 1891, the next 
to it 3°-4 in March 1891, and 4°-2 on several days in March 1890. 
The range of surface temperature is the greatest observed, being 12°'5. 
The bottom temperature had a maximum of 14°*5 on August 8, 1889, 
and a minimum of 3°-l on March 11, 1891, giving a range of 11° '4. 
The curve for monthly mean surface temperature during 1890 is the 
most symmetrical of any represented in Diagram IV., showing minima 
in the end of winter (January 1890, February, and possibly March 
1891), and a maximum in September (the maximum in 1889 was in 
August). The fall of temperature was much more rapid than the 
rise ; as much heat beiiig lost in three months as had been gained 
in five. The ten-day means given in Table I. are deserving of careful 
study, and show many relations of an interesting kind. When the 
compensating disturbances are allowed for, the monotonous state of 
matters that a glance at the mean temperatures might suggest is 
found to be very far from correct. 
