would materially change the concepts of the phylogeny 

 and classification presented here. 



For each of the 10 families treated here, one relatively 

 generalized species is described in detail and extensively 

 illustrated, with other species as appropriate described 

 and illustrated less fully, except in the Tetraodontidae, 

 with two species representative of the two subfamilies 

 that have often been considered as separate families 

 being given the fuller descriptive and illustrative treat- 

 ment to aid in comparisons of their osteological distinc- 

 tiveness. Line drawings show the most diagnostically 

 important or anatomically interesting external features 

 of representatives of each group. 



Lateral view illustrations of the entire skeleton are 

 usually supplemented by other views of various parts of 

 the skeleton showing features of particular phylogenetic 

 or morphological interest. 



Statements concerning the numbers of such serial 

 elements as basal pterygiophores, epipleurals, neural 

 and haemal spines, epurals, and hypurals are based 

 mostly on cleared and stained specimens and not on 

 radiographs, because all of the members of such series of 

 these elements usually are not clearly defined in radio- 

 graphs due to the often thickened nature of the scales. 

 Vertebral counts are based on both cleared and stained 

 specimens and radiographs, except for those of diodon- 

 tids in which radiographs of the vertebral column are 

 usually undeciferable because of the massive roots of the 

 skin spines. 



The lateral view illustrations of entire skeletons and of 

 most of the skeletal parts were prepared by photo- 

 graphing the skeleton or its parts submerged in a gly- 

 cerin solution. The 35 mm negatives were projected and 

 outline drawings showing as much detail as possible 

 made, of about 40 to 50 cm length in the case of entire 

 skeletons. Details were filled in by examination through 

 a dissecting microscope of the photographed skeleton or 

 part. The completed pencil drawings were transferred by 

 ink tracing on paper, with contours and depth approxi- 

 mately shown by varying intensities of stippling. While 

 the pencil drawings and ink tracings are all by the 

 author, the stippling is by a combination of the author 

 and his illustrators listed in the Acknowledgments. 

 Photographed entire specimens and/or one or more ad- 

 ditional specimens of the same species were partially 

 disarticulated in order to prepare, if necessary, drawings 

 of selected parts or views of regions of diagnostic or mor- 

 phological interest, as well as generally to examine the 

 skeleton and its parts in greater detail. A few of the sim- 

 pler supplemental illustrations were made with the aid of 

 proportional dividers rather than photography. Some of 

 the drawings are composite, based on several specimens, 

 and are so indicated. 



Fin rays usually are shown only partially and diagram- 

 matically as an outline of the positions of the bases of the 

 rays, with, for the caudal fin, the unbranched rays in- 

 dicated by solid bases and the branched rays by open 

 bases. Only the pelvic fin rays and uppermost pectoral 

 fin ray routinely were fully drawn. Epipleurals, especial- 

 ly if slender, usually are shown in solid black rather than 



stipple. The finer details of the surface sculpturing of the 

 bones usually is not shown, for, even though it is beauti- 

 ful, it is exceedingly time consuming to accurately por- 

 tray and normally is of little value to an understanding of 

 the relationships of the species in question. 



The otoliths of the species treated here are not illus- 

 trated, for they only rarely were intact and uncorroded 

 enough in the cleared and stained study material to be 

 useful. 



The osteological descriptions of the species treated in 

 greatest detail were written from microscopic reexamina- 

 tion of the skeletons or parts thereof, and corrections 

 were made in the drawings whenever necessary. Correc- 

 tions were frequent, for nothing brings to light the errors 

 in illustrations more quickly than having to verbally 

 describe the shapes of bones and, particularly, their ar- 

 ticulations. 



The osteological terminology used here is conserva- 

 tive, and hopefully that most readily understandable by 

 the majority of ichthyologists. For the most part, it is 

 that of Starks (1901), although such terms as ptero- 

 sphenoid, ectopterygoid, first pharyngobranchial, 

 cleithrum, and scapula are substituted, respectively, for 

 Starks alisphenoid, pterygoid, suspensory pharyngeal, 

 clavicle, and hypocoracoid. Certain names, such as 

 prefrontal rather than lateral ethmoid or dermethmoid, 

 are used in order to be deliberately topological and to 

 avoid implications of the dermal versus endochondral 

 origin of the bone. 



Subsequent to the beginning of this monograph, and of 

 the pervasive decisions on the names of bones to be 

 employed in the text and illustrations, several 

 researchers have provided more precise and homologous- 

 ly correct terminologies for certain skeletal regions than 

 used here. It has not been possible for reasons of prac- 

 ticality to make the numerous desirable changes, es- 

 pecially in labeling of the illustrations, that would be 

 necessitated by incorporating these changing ter- 

 minologies into the present work. Examples of the more 

 modem and accurate names for certain skeletal regions, 

 such as the branchial arches and lower jaw, are found in 

 the exemplary publications of Weitzman (e.g., 1967 et 

 seq.) and Nelson (e.g., 1969 et seq.). 



In describing the types of articulations between bones, 

 purely descriptive phrases often are used, for the tech- 

 nical terminology is not standardized and it is sometimes 

 more precise than my observations. Whenever an 

 amount of cartilage observable under the dissecting 

 microscope (ca. 30x ) is seen to intervene between the ar- 

 ticular faces of two bones, these bones are said to ar- 

 ticulate through cartilage. When bones have relatively 

 smooth articular faces held to one another without the 

 intervention of cartilage, they are said to articulate by 

 fibrous tissue. A large number of bones £ire said to ar- 

 ticulate with one another by interdigitation, and this 

 simply means that the closely apposed surfaces of two 

 bones have their articulation strengthened by delicate to 

 coarse emarginations of one bone fitting into similar in- 

 dentations of the other bone. At its fullest development, 

 this interdigitation can be distinguished from fusion only 



