136 ROBERT TRACY JACKSON ON ECHINI. 



They are: two specimens with oculars I, II insert; two, I, IV; thi-ee, V, IV; two, V, II; six- 

 teen, I, V, II (text-fig. 145); and one with I, IV, II insert and V, III exsert. In this last case 

 genitals 4, 5 are fused (text-fig. 147), shutting out ocular V, so that it is evidently a case of 

 incomplete I, V, IV, II. It is the only instance found of just this combination in any sea-urchin. 



At Dumpling Islands this species attains an exce]otionally large size, and the next two 

 sizes could not be collected often at other localities with which I am acquainted. The series 

 60 to 70 mm. in diameter is a small one (121 specimens) compared with the preceding. One 

 specimen, 0.8%, has ocular I only insert, 95% have I, V, and two specimens, 1.6%, have I, 

 V, IV insert. The second specimen was found in the last few examined. This is mentioned 

 as, if by chance it had not been found, the percentage of progressive variants would have been 

 much less, smaller even than in specimens 20 to 30 mm. in diameter. The generally expressed 

 view that more oculars are insert in large specimens is quite without foundation in any species 

 as far as my observations go. The aberrant variants of this series are 2%. Two specimens 

 have oculars I, II insert (text-fig. 141), and one has I, V, II insert. The last series, 70 to 75 

 mm. in diameter (seven specimens), has 100 % with oculars I, V insert. In a larger series prob- 

 ably some variants would have been found, but as far as it goes, it emphasizes the fact borne 

 out by other species that the largest specimens have the species character, and that variants, 

 both arrested and progressive, are more frequent in smaller individuals. 



This large series from Dumpling Islands demonstrates that very young specimens show 

 the passage from all exsert to ocular I insert and then to the bivium insert in ontogeny; older 

 but yet young specimens show that progressive variants maj^ appear even to the full specific 

 proportion before the full percentage of the specific character is established. Specimens half- 

 grown (30 mm. or more) have about the same percentage and range of the arrested, typical, 

 and progressive characters as in fully adult specimens. Large individuals have no greater 

 number of oculars insert than smaller individuals, excluding youthful stages. 



Through the kind help of friends and from personal collecting, series of Strongylocentrotus 

 drobachiensis have been obtained from a number of localities for study, to see what the specific 

 range might be. In all localities the character is for oculars I, V to be insert, but the percentage 

 of this character varies with locality, and also the relative proportion of arrested and progressive 

 variants varies very widely. The localities represented are arranged in the table on the basis 

 of these characters (p. 143). 



The most primitive type of the species seen is from Gullmar Fjord on the Swedish coast of 

 Skagerrak. From this locality Dr. Mortensen kindly sent me fifty specimens. On inquiry 

 he informs me that this fjord receives a considerable volume of fresh water from a river, so that 

 the water is at times distinctly brackish. The specimens are medium sized individuals. Thirty 

 percent have ocular I insert and only 58 % oculars I, V insert. There are no typical progressive 

 variants, but a high percentage of aberrants, 12%. Of these, one specimen has oculars I, II, 



