Luccrnaria quadricoinis and rdated Species. 22'.\ 



inorpholof^iciil cliaractcr like tlio ctclciitcronol' tlie stalk heiiii^- 

 siii<;le- (^LuceriKiria) or fuiu -chain bcied [Ilalivlyslus) — espe- 

 cially as variations, numeral and other, so frequently occur in 

 Luccniariiin tentacles (Clark, Beaumont, Hornell, lirowne). 

 Antipa (181)2) has drscribud a young Craterolophns tetliys 

 ■which had a single tentacle in each of four neighbouring 

 octants. Browne (18it5, p. 4) mentions two IJaHchjstus 

 octoradiatus having "capitate tentacles on the margin ot the 

 umbrella in an abnormal position." Further, Hornell (ISUo, 

 p. 208) has noted the presence of "marginal bodies" in the 

 young of L, cainpaiiulata. Variations of this type must, 1 

 think, be regardet-l as "vestiges of the tentaculocyst-rudiments 

 of ancestral scyphistomata " (Hurst, p. 2.14) ; it is note- 

 worthy that these vestigial characters occur chiefly in young 

 specimens, just as the presence of tentacles on the marginal 

 anchors is a normal condition of young HaUcli/i<tus (Beau- 

 mont, 1900). 



These Clyde specimens I regard as young Lucern<iria 

 quadricornisy Miiller, and they agree with that species in the 

 following chaiaeters : — Funnel-shaped, slightly four-sided ; 

 stalk single-chambered, from equal to to twice the length of 

 the body, cylindrical, annular in contraction ; perradial bays 

 twice as broad and deep as the interradial ; gonads extending 

 to the ends of the arms ; arms ending in a cluster of ca[)itate 

 tentacles, eight to twenly-one in number. 



The smaller Clyde specimens were yellow in colour, like 

 the Port Erin ones ; the larger were olivaceous brown, like 

 the Shetland one. 



These specimens also are, I think, referable to L. quadri- 

 coriiis, Midler. The structure of the gonads is very siniilar 

 to that of Ihdiclijstus (Clark, 1878, p. 07). They form eight 

 udradial bands, composed of " hollow s{)heroidal saccules . . . 

 attached to the inner taces of the circumoral parietes. They 

 are totally disconnected from each other, but usually so 

 crowded that their peripheries come in contact and mutually 

 mould themselves into polygonal shapes." The saccules 

 open, each by a short oviduct, into the radial pouches. The 

 specimen which Beaumont sectioned was ripe, but rather 

 crumpled, which makes the details difficult to follow; but, 

 after a careful comparison, I think they agree with a C^lyde 

 specimen (I'Jl'J. vii. 11) which is as described. 



My thanks are due to the Director of the IMymouth Labo- 

 ratory for his courtesy in lending Mr. Beaumont's sections 

 to me. 



