446 MR. HERBERT L. HAWKIXS OX 



group becomes an almost impossible task. The features which 

 seem of essential importance in one genus may be quite absent or 

 profoundly modified in another. But in the case of the 

 Holectypoida, the length of time during which this plasticity of 

 structure was retained gives possibility for a classification that is 

 fairly in accord with the evolution of the group, and at the same 

 time is free from a confusing multiplicity of detail. However, 

 as will be seen on a comparison of the scheme submitted below 

 with the genealogical table given in a later part of the paper, the 

 two groupings do not agree in every particular, AH of th© 

 genera have a great phylogenetic significance, and would, if the 

 classification were to do justice to that importance, require each 

 a separate family. Probably, as our knowledge of the relations 

 of the group extends, the present genera will become the bases 

 of distinct families, and will be themselves divided into many 

 smaller sections. The great variety of species which are at 

 present groujied under the generic names Holectjipus and Discoidea 

 seems to lend support to this belief. For convenience of reference, 

 it has seemed preferable to retain, as far as possible, a more 

 generalized system of grouping than, by comparison with other 

 orders of Bchinoidea, the individual peculiarities of the genera 

 in reality demand. 



The Holectypoida are an order intermediate in characters 

 between the Echinoidea Regularia and all the various orders of 

 the Irregularia. The features in their structure which are 

 naturally the most uniform in character are therefore those 

 relics of " Regularity " that they retain. They show a persistently 

 retarded progress in their evolution, and from them, at various 

 points, relatively accelerated oft'shoots break free. Throughout 

 the entire group two features remain constant in their presence, 

 although they undergo a gradual reduction in the perfection of 

 their development. These features are :— the existence of a 

 masticatory apparatus ; and the presence of external peristomial 

 branchiae. Both of these structures are of essential importance 

 physiologically, and both fortunately leave traces of their existence 

 on the skeletal structures. It is a postulate (which probably 

 expresses a fact) that, whenever the peristome is centrally situated, 

 and at the same time circular in outline, jaws are present. 

 Certainly the existence of a well -developed perignathic girdle 

 indicates their presence, so that, when either of these characters 

 can be observed, the existence of a masticatory apparatus can be 

 infeiTed, even if it has not been discovered. The branchial 

 incisions on the peristome margin are naturally easy to recognize 

 when that part of the test is preserved. 



After the scheme of classification has been summarized, revised 

 diagnoses will be given for ^ the order, families, and genera. 

 SubseqiTently, the reasons for the exclusion of some genei'a which 

 have hitherto been classed with the Holectypoida will be 

 discussed, and their positions in the other orders indicated. 



