80 BULLETIN OF THE BUREAU OF FISHERIES 



invariably been lowered as near to bottom as feasible, so as to sample the whole column 

 of water. As yet we have not attempted a quantitative survey of any particular 

 stratum, though, from the nature of the case, the hauls in the shallow coastal zone 

 have been confined to a thin layer of water. 



The results of the vertical hauls are supplemented by the much more numerous 

 horizontal hauls, made with various nets and covering the gulf generally at most 

 seasons of the year. Inasmuch as the quantitative value of horizontal hauls has 

 often been disputed, I must admit at once that they seldom fulfill the basic requirement 

 of fishing through a column of water of known length. Furthermore, while the level 

 at which an ordinary open net works for the major part of the haul can be determined 

 within reasonable limits if it is used at moderate depths, its yield can not be depended 

 upon as an index of the richness of the plankton at that particular depth unless cor- 

 roborated by other evidence, because it may have passed through a swarm of copepods 

 or what not on its way up or down. Horizontal hauls made in deep water, say of 

 500 meters or more, have little quantitative value if of short duration, because the 

 horizontal journey made by the net may then be little if any longer than the vertical, 

 which, of course, may be equally true of individual hauls in shallow water under 

 exceptional circumstances. In general, however, it is safe to assume that when the 

 horizontal distance through which the net works exceeds the vertical manyfold, as 

 is the case for shallow hauls of considerable duration (for example, our standard of 

 half an hour at 100 meters or shallower), considerable weight may be given to the 

 average quantitative results of several hauls, the more so the greater the discrepancy 

 between their horizontal and vertical portions, hauls at the surface being entirely 

 satisfactory in this respect. In short, while everyone agrees that it is idle and 

 misleading to expect precise quantitative data from ordinary tow nets used hori- 

 zontally from a moving vessel, there is no need of going to the other extreme, as 

 some students have done, and discarding a method that is not only so convenient but 

 so often available when rough weather prohibits vertical hauls. 37 As a matter of 

 fact, if they are interpreted with common sense and made at appropriate levels in 

 the water, the catches of the horizontal tow nets often throw much light on the quan- 

 titative distribution of the animal plankton, especially in preliminary surveys. At 

 the worst they can be trusted to reveal the existence of areas of markedly rich or of 

 very scanty plankton, for no one can deny that the plankton must be more abundant 

 where tows are uniformly productive than where the same nets as regularly yield 

 little or nothing, especially at times and places when and where the larger animals 

 occur in local shoals, which the vertical net may miss altogether but which a long 

 horizontal tow is almost certain to encounter. 



Thus, to quote only one example, Jespersen (1924) was able to demonstrate very 

 wide differences in the abundance of zooplankton in different parts of the Atlantic, 

 from horizontal hauls of long duration with large nets, especially the general poverty 

 of the so-called "Sargasso Sea." 



•' An excellent example of the light which horizontal hauls may throw on the fluctuating abundance of the plankton is afforded 

 by the long-continued series of tow nettings carried out by the Marine Biological Laboratory at Port Erin, on the Isle of Man, 

 under Professor Herdman's direction. 



