for the genera and species studied. Kiausewitz (1964) ac- 

 curately described and illustrated the skull of an 

 Arothron, and Rishi (1969) that of a Chelonodon. 



The otic region of the skull of plectognaths has been 

 the subject of much misinterpretation in the literature. 

 One source of the error has evidently been the feeling 

 that the plectognaths, like most other fishes, should 

 possess a pair of parietals. But the plectognaths have lost 

 all trace of the parietals and what have been called the 

 parietals in plectognaths are portions of the epiotics, 

 sphenotics, or frontals. Wellenbergh (1840) referred to 

 a portion of the epiotic in Mola as a parietal. Bruhl (1856) 

 described the parietals as being absent in all plectog- 

 naths except Diodon and Mola, calling the posterior 

 part of the sphenotic in Diodon a parietal and accepting 

 Wellenbergh's statement that a parietal existed in Mola. 

 In HoUard's (1853, 1854a, b, 1855, 1857a, b, 1860) various 

 publications the posterior portion of the frontal is called 

 the parietal, although a definite suture between the 

 bones is not usually shown in his figures. Goeldi (1884) 

 pointed out that many of the sutures shown by Hollard 

 were artifacts and criticized Hollard for the excessive 

 number of bones that had been described as separate ele- 

 ments. What Siebenrock (1901) described as a parietal in 

 Balistes, even though he could not find one in Tria- 

 canthus, or Monacanthus, is probably a part of the 

 epiotic. Supino (1905) thought that in Balistes the 

 parietals were fused to the frontals, but in Mola he 

 described the medial half of the pterotic as being a 

 parietal. Kaschkaroff (1914a) described a parietal as be- 

 ing present only in Balistes, his parietal evidently being 

 part of the epiotic. Kaschkaroff believed that the frontal 

 of Mola might represent the product of the fusion of the 

 frontal with the parietal. Rosen (1916a:13) said that the 

 parietals were absent in plectognaths, and expressed the 

 prevailing view that "they have perhaps fused with the 

 frontals." The above statements are exclusive of those in- 

 stances in which there is a purely nomenclatural dif- 

 ference, e.g., Regan's (1903a) use of the term "parietal" 

 as being synonymous with "epiotic." Since Rosen's time, 

 no knowledgeable worker has described a parietal in a 

 plectognath fish, and if that bone has fused with the 

 frontal it has done so indistinguishably. Not even a trace 

 of a separate ossification which could conceivably be 

 referred to as a parietal has been seen in any of the 

 developmental stages of various plectognaths examined 

 for this work. Under these circumstances it would seem 

 best to assume that the parietal simply fails to develop in 

 plectognaths. 



The other major source of error in the interpretation of 

 the otic region has been the description of the upper and 

 lower surfaces of the epiotic as two separate and distinct 

 bones, because of a certain anatomical peculiarity of the 

 development of the endochondral bones, particularly in 

 the otic region. The otic bones begin to develop in the 

 normal manner as small ossifications in the appropriate 

 region of the chondrocranium. From this center of os- 

 sification in a plate of cartilage, the gradual peripheral 

 movement of the ossification does not proceed evenly. 

 Rather, the ossification spreads through the upper sur- 



face and through the lower surface of the cartilage, so 

 that toward its edges the bone is composed of an upper 

 and a lower layer separated by cartilage. The two layers 

 are continuous with one another only at their original 

 center of ossification. As the specimen becomes larger, 

 there is a slow and gradual ossification that fills in the 

 otherwise cartilaginous area between the two peripheral 

 layers of the bone. It is only in extremely large specimens 

 that all of the cartilage between the double-layered 

 peripheral parts of the bone is replaced by ossified tis- 

 sue. Thus, in the descriptions given here of the otic and 

 other regions, there is reference to these endochondral 

 bones having the edges filled with cartilage, simply 

 because in all but the largest specimens of any par- 

 ticular species the replacement of the cartilage between 

 the upper and lower layers in the peripheral region of the 

 bone is incomplete. This double-layered condition is par- 

 ticularly evident in the epiotic and sphenotic, and is 

 progressively less evident in the pterotic, prootic, exoc- 

 cipital, supraoccipital, and the other endochondral bones 

 of the skull. In adult specimens the cartilage-filled edges 

 of these endochondral bones are usually not evident on 

 external examination, and they can be seen only when 

 the bones are disarticulated. 



The reason that this condition is not particularly ap- 

 parent is that the edges of the upper layers of two ad- 

 jacent bones, e.g., epiotic and sphenotic, may inter- 

 digitate with one another, hiding the fact that just below 

 this interdigitated surface there is a thin layer of car- 

 tilage separating the upper and lower layers of the bones. 

 It is usually true, except in large adults, that when the 

 upper layers of two bones articulate by interdigitation, 

 the lower layers of these bones are separated from each 

 other by a small amount of cartilage, this cartilage ex- 

 tending for a short distance into the substance of each of 

 the bones. It must be emphasized that even though a 

 bone such as the epiotic has an upper and a lower layer 

 separated by cartilage peripherally, these two layers 

 converge toward one another centrally to be connected 

 by a core of bone representing the original center of ossifi- 

 cation. This central core of bone is delicate enough in 

 some cases to be easily broken when skeletal material is 

 prepared by drying, because of the contorted shrinkage 

 of the cartilage that lies between the double-layered 

 peripheral regions of the bone. If the drying of the skele- 

 ton does not break the central core, then the process of 

 disarticulation of the skull often will. 



It is thus not surprising to find that Klein (1872, 1881, 

 1884, 1885, 1886) stated that in plectognaths there is an 

 epiotic which is covered over by a parietal. Klein's 

 "parietal" was simply the upper layer of the epiotic, 

 which had become separated from its lower layer during 

 the drying and disarticulation of the skull by breaking of 

 the central ossified core which normally makes the two 

 layers continuous. Awati and Bal (1933) have done essen- 

 tially the same thing in their study oiFugu oblongus, for 

 they refer to the ventral surface of the epiotic as the 

 "epiotic," but to its dorsal surface as the "parietal." 

 Furthermore, they refer to the ventral surface of the 

 sphenotic as the "sphenotic," but to its dorsal surface as 



