structing - a system taking regard to botk, and thereafter one author 
bases tbe system on the medusæ, another on the hydroids. I 
think the system must be based on the hydroid polyps, because they 
are undoubtedly the original forms, from the sporosacs of whicli 
medusæ have derived later; on the other hånd, regard must also 
be taken to the medusæ. When hydroids, which are considered to 
be nearly related species, are said to produce medusæ belonging to 
quite different families, or vice versa, we must believe, that any 
mistake or misunderstanding of any important systematic character 
must have taken place. I feel sure that a natural common classi- 
fication of hydroids and medusæ must be possible, and I think the 
difficulties and the apparent contradictions are due partly to our 
great ignorance as to which hydroids and wbich medusæ belong to 
each other, partly to the disagreements as to which characters in 
the medusæ are of systematic importance. Some authors lay stress 
on the sense organs, others attach importance to the number and 
shape of the tentacles or the radial canals, on the presence or ab¬ 
sence of cirri and sensory clubs and so on. Therefore, within the 
medusæ, forms as Staurophora, Laodice, Tiaropsis etc. have been 
very much thrown around in the system. 
Not less difficult is the question, whether the one or the other 
main-group of u Campanulinidæ" is related to the Lafoeidæ. The 
medusæ cannot help us in this respect, as we do not know any 
species of Lafoeidæ producing medusæ. A. Agassiz (1865) thought, 
from his experiments in aquarium, that Laodice calcarata was the 
medusa corresponding to a small species of Lafoeidæ , Hebella 
(Lafoea) calcarata Ag. Browne (1907) has criticized Agassiz’s 
statement and has very much doubt as to the connection between 
the medusa and the hydroid in question. Other species of Laodice 
are known to belong to Cuspidella. — L e v i n s e n finds several 
resemblances between genera and species of Lafoeidæ and genera 
and species of both main-groups of u Campanulinidæ'. Calycella 
syringa and Tetrapoma quadridentatum present a great likeness to 
Lafoéa pocillum and L. pygmæa with their carapanulate-tubular 
