CEPHALOPODA OF THE HAWAIIAN ISLANDS. 269 



or apology even to the layman. This is particularly needful concerning the Octopoda, 

 as in this group it is often very difficult to lay down on paper a hard and fast line respect- 

 ing a given character which will invariably serve to separate a species from some other 

 perhaps very closely allied. Here, in the absence of direct comparison between speci- 

 mens, a sort of average of the entire physiognomy is largely depended upon, taking into 

 consideration the relative length of the arms, the shape and extent of the umbrella, 

 the presence or absence of ornamental processes of various sorts upon the integument, 

 and the more evident conditions of preservation, which frequently affect the features 

 already stated to a degree which can scarcely be too strongly emphasized. When the 

 specimens are males, however, the structure of the hectocotylized arm is variously 

 modified, usually constant in its peculiarities, and hence a criterion of the highest import- 

 ance; perhaps no other single feature so well maintains its value in preserved material. 

 I am also inclined to treat with respect any decided peculiarities of color, particularly 

 when the various pigments appear to be disposed after the manner of a definite pattern. 

 Frequently the so-called "funnel organ" exhibits tangible modifications in shape, though 

 whether these latter are more valuable than confusing as a key to interrelationship is not 

 yet apparent. 



Among the Decapoda some of the above-mentioned features are of very minor 

 specific significance. Here I have frequently given considerable weight to relatively 

 minute differences in the form and arrangement of the suckers (or hooks) on the sessile 

 arms and more particularly, on the terminal clubs of the tentacles, as these characters, 

 even though small, are definitely to be apprehended and usually little affected by the 

 action of the preserving medium. Nevertheless, there is certainly need for a greater 

 quantity of comparative data showing the range of variation in these organs and until 

 this is available there is constant danger that too much emphasis has been placed upon 

 their details. Among the luminous forms, particularly the CEgopsida, the arrangement 

 and structure of the photogenic organs is nearly always subject to important modification, 

 both specific and generic or even of higher significance. The funnel organ appears to 

 be of less practical systematic value in the Decapoda than in the Octopoda, for although 

 it undergoes considerable modification among different genera (particularly the various 

 Cranchiidae), the differences between those of closely allied species are apt to be too 

 slight to be appreciable. 



On account of their easy preservation as "hard parts," the gladius and radula have 

 occupied a dominant position in the schemes of classification of many of the older authors 

 as well as a few of the more recent ones, but the undesirable mutilation of the specimen 

 entailed by their examination, coupled with the writer's belief that their relative import- 

 ance has been greatly overrated, has in the present work prevented their receiving the 

 attention to which they are perhaps more justly entitled. Other "hard parts" which 

 are ofttimes very useful to observe are the horny rings which arm the apertures of the 

 suckers in most if not all Decapoda and which often show interesting modifications, 

 chiefly depending upon the varying degrees of smoothness or denticulation of their outer 

 margins. 



17311°— 14 2 



