1H4 D. J. HCOURFIKI.I) OX hJNTUM* (JriTKAt'A. 



fact that in this particular area some of the species iiave 

 appeared and disappeared each year with very considerable 

 regularity is a point that must have at least some significance 

 in relation to the life-histories of these forms and the conditions 

 necessary for their existence. Leaving out of consideration the 

 rarer forms, whose capture must have been more or less acci- 

 dental, it appears that about sixteen species of Cladocera, four 

 of Ostracoda, and three or four of Copepoda show a limited 

 seasonal distribution. These, with their periods of activity, 

 are shown in table on p. 165. 



It is true that b}' looking only to the earliest and latest 

 records obtained during the three years, as done in the following 

 list, the active period seems to extend in some cases to as much 

 as nine or ten months, but if taken year by year it will generally 

 be found somewhat shorter ; and I have also observed that in 

 some of these instances of exceptionally early or late records, 

 onl}' a single specimen was seen or only young forms taken. 

 On the other hand it will be seen that in two cases the active 

 period only lasts for three months. A few further comments 

 on this subject w^ill be given later among the detailed notes of 

 species forming the second part of this paper. 



The next result, and the last that seems at all definite, is in 

 close dependence upon the periods of activity and rest alluded 

 to in the preceding paragraph. It is that a w^ell-defined cycle 

 of seasonal distribution of species has been made out in the 

 case of the Cladocera. The evidence of each year on this point 

 is practically the same, as show^n by the following statement of 

 the number of species of Cladocera taken at each visit : — 



Jan. Feb. Mar. Apl. May Jane Jnly Aug. Sept. 



1890 ... — 6 6 11 16 — 18 13 22 



1891 ... 1 7 6 10 9 — — 17 18 



1892 ... 6 — 5 11 11 16 17 — 21 



Averagef 4 7 6 11 12 16 18 15 20 18 17 10 



It will be seen that, with the exception of two or three 

 slight breaks, there is a gradual increase during each year 

 from January to September, then a slight decrease to about 



* This and her records for December, 1892, have, of course, been in- 

 serted since thc^ *-ding of the paper. 



+ Fractions ^{^,e-half and two-thirds counted as one in this and other 

 uverayes given t . ucdlIv . 



