72 



in which the helerozooecia are either constantly occturing or always absent anil 

 in that case their appearance or absence will help to characterize the repective 

 families or genera. For instance the absence of avicularia is to such a degree 

 bound up with the notion Electra, that the discovery of an avicuhiria-bearing /i/ec/ra 

 species would rightly be looked upon as very remarkable. We may mention the 

 Adeonidae, Catenariidae, Celleporidae, Holoporellidite and Tbalamoporellidac as ex- 

 amples of families with constantly occurring avicularia, while these are absent 

 in the Aeleidae and Steyanoj)(>rellid(ie. They appear for instance constantly in the 

 genera Callopord and Exuchella, while they are absent in the genera Membrani- 

 poru (s. str.) and Electra. 



The oci'cia present a similar inconstancy in their occurrence to the hetero- 

 zoo'cia, as they are (juite absent in a number of families (e. g. Adeonidae. Slegano- 

 porellidae and Aeteidcie), genera (e. g. He<ini(t, Meiubraitiporu (s. st.) Cupid<tri(i and 

 Lunularid) and species, and in many sjiecies they ap])ear very inconstantly and 

 by no means in all the colonies. For instance, in Discopora uerrucosa we only 

 find ocecia in the colonics from deeper water, never in coastal forms. There- 

 fore a genus or species cannot be based only on the ditlerence, that they have 

 or are without ooccia, whereas a constant occurrence of these formations can be 

 used as an auxiliary character. What has been said about the systematic value 

 of the fact, that heterozocecia or ooecia occur in a species, genus or family, does 

 not exclude the different structure of the heterozocecia and the otecia from having 

 a great systematic importance and we shall discuss this matter further in the 

 following sections. 



With regard to the kenozooecia, their systematic impDrtance is very diverse, 

 and the small triangular spaces for instance, which appear between the zou'cia 

 in different forms, are not always constant in the species. This applies for ex- 

 ample to Membranipora Lacroixi. That the so-called radical fibres, which serve to 

 fasten freely growing colonies of less solid materials, only have a very slight system- 

 atic importance, is evident, partly from Ihe fact that they are found in so many 

 families (c. g. Flustridae, Bicellariidae, Cellidariidae, Scrnpocellariidde, Cotenariidae 

 etc.), partly because their occurrence is dependent on the free condition of the 

 colony, which has no systematic importance. We may give here a few examples 

 to show that the radical fibres can be present or absent in forms of growth of 

 the same species as well as within closely related species. Such radical fibres 

 can for example be found in Stej/anoporella neozekinica, which occurs in pillar- 

 shaped trunks, while they are absent in an incrusting form, which Harmer 

 considers as a variety, var. magnified, of the same species. A similar relation is 

 found between the freely growing species: Microporella fldbellaris and Mic. margi- 



