FISHERY BULLETIN: VOL. 74, NO. 3 



not be relevant to present purposes and would 

 carry away from the intent of this communication, 

 although the equation of Breder (1954) should be 

 useful to such a study. 



The main thrusts of the students of traffic flow 

 have been concerned with such things as problems 

 of delays, queueing, road junction, traffic signals, 

 analogies to fluid movements, and follow-the- 

 leader sequences. 



The study of fish schools has not yet reached into 

 these matters, although they all bear a one-to-one 

 resemblance to similar items in schools. This 

 undeveloped area is difl^icult to enter into deeply 

 partly because there is no facile way to keep track 

 of each individual. The analysis of the behavior of 

 individuals in a school based on the data of Hunter 

 (1966) could be considered as a start in this 

 direction. 



Influence of Body Forms 



There is a marked positive relationship between 

 schooling and the extent of streamlining of the 

 general contours and of the drag-reducing surface 

 details of fishes that, in the most advanced ob- 

 ligates, can be considered exquisite. Parallel to 

 this is an equally marked negative relationship 

 between schooling and special surface features of 

 the eruptive sort. At this end of the series, the 

 fishes are not schoolers at all, nor even aggrega- 

 tors, but are usually solitary, neutral, or agonistic 

 toward their fellows. All of this can be shown to be 

 related to mechanistic details covering the manner 

 of life of the individuals involved. 



For example, we know of no obligate schoolers 

 such as clupeids or scombroids that have any 

 drag-producing extensions, while the vast majori- 

 ty show beautiful fairing even in the manner that 

 the maxillary fits into a matching recess when the 

 mouth is closed and in the slot that the depressed 

 dorsal fin fits into as shown in Scomberomorus. 

 Such niceties are not to be found in the facultative 

 schoolers such as most of the Salmonidae, Cyprin- 

 idae, and Serranidae. In the essentially non- 

 schooling fishes, the streamlining often becomes 

 less effective and outgrowths from the integu- 

 ment and eruptive structures become more and 

 more extreme as in Hippocampus, the Scorpaen- 

 idae, Cyclopteridae, and Diodontidae. With this 

 comes slower swimming speeds and an increasing 

 tendency to reduce swimming to a minor roll, as in 

 some of the Scorpaenidae and all of the 

 Antennariidae. 



The remainder of fishes to be considered here 

 are those that show a depth^^ equal to or greater 

 than their lengths. These are often facultative 

 schoolers. Families in which this is a usual or 

 frequent condition include the Stromateidae, 

 Ephippidae (including the extreme platacids), 

 Chaetodontidae, and Acanthuridae. Many others 

 show an approach to the condition, as in the 

 Pomacentridae. In addition to these, there are a 

 considerable number of families in which one or a 

 few species have the necessary characteristics, as 

 the Carangidae and Cichlidae. 



The schools that are formed by fishes of great 

 body depth are superficially very similar to those 

 formed by fishes with fusiform outlines. A school 

 of deep-bodied fishes is, however, automatically 

 tighter because the greater depth of body intrudes 

 into the swimming areas of the layer of fishes 

 above as well as the layer below. 



There is both mechanical and hydrodynamic 

 interference and an optical occlusion that is much 

 more severe because of the greater area of the 

 sides of these fishes. This leads to greater difficulty 

 in making sharp turns. These conditions can only 

 be relieved by loosening the school in the vertical 

 direction. How much mutual swimming facilita- 

 tion is lost by this loosening is not known. Figure 

 19 illustrates these conditions with a head-on 

 photograph of an extremely loose school of Chae- 

 todipterusfaber (Broussonet). 



The only other fishes known to form schools are 

 those in which the longitudinal axes do not lie 

 parallel to their line of travel. They include various 

 characins, the "head standers" of aquarists, and 

 some aulostomoids, the best known of which are 

 Aeoliscus and Macrorhamphosus. These evidently 

 swim with the head up or down (Atz 1962, 

 Klausewitz 1963). There is no data on any aspect of 

 their hydrodynamics nor on their mucus. These 

 forms, therefore, are not discussed here. 



Sizes of Fishes in a School. 



The variation in the lengths of individuals in a 

 school usually reaches no more than 30%. The 

 difference between the length of the largest fish 

 minus that of the smallest fish in a given school is 

 expressed as a percentage in this notation. Data 

 from Breder (1954), recalculated for present pur- 



'^This is not the body depth of taxonomists, but the vertical 

 depth of the entire profile, including the extent of the dorsal and 

 anal fin in that dimension. 



488 



