ALTERNATION OF GENERATIONS. 123 



the umbrella by an investing sac, in which the canals are 

 imperfect or altogether wanting. 



Even in allied species, it is to be noticed, that there is a 

 great diversity in the development of these structures. 

 Thus in Laomedea geniculata the reproductive elements are 

 contained in free swimming mesdusoids with marginal 

 cirrhi, and a central proboscidiform mouth, round the at- 

 tachment of which the ovarian chambers are situated. In 

 Laomedea Loveni* they assume the form of an inflated 

 bell surrounding the ovary, and fringed at its free edge with 

 minute tentacles ; and they remain during their brief period 

 of life attached to the edge of the homy " ovigerous cap- 

 sule" characteristic of the group, and there emit the sper- 

 matozoa or ova with which they were charged ; after which 

 they wither like blossoms, to be succeeded by a new expan- 

 sion. In Laomedea lacerata the ovarian sac advances to 

 the mouth of the capsule, but instead of a bell-shaped en- 

 velope becomes invested merely by a thick gelatinous coat, -f 

 Even in the same species there is sometimes as great a 

 diversity in the opposite sexes ; thus in L, geniculata the 

 ova are formed in free swimming medusoids, the sperma- 

 tozoa in simple cysts, permanently attached.^ 



Professor Allman refers to the reproductive sacs of 

 Tubular ia indimsa, which never become detached, as of 



* Allman in Annals of Nat. Hist., 3d Ser., IV., p. 140 the same 

 species apparently which Dr. Wright refers to as confounded by Johnston 

 with another having free medusoids, under the name of L. dichotoma. 

 (Edin. Philos. Journal, Jany, 1859, p. 110). This form of extracapsular 

 medusoid, Allman terms meconidium, from its resemblance to a poppy- 

 head. 



f Dr. Wright, Op. Cit. 



$ At least Professor Schultze is quoted by Dr. Wright (Edinburgh 

 Philosophical Journal, July, 1856, p. 85, note) as having discovered 

 sperm capsules in this zoophyte, but there is a degree of confusion among 

 the species which somewhat invalidates the conclusion in the text. 



G 2 



