21 



the pool to the nearest main ditch, and not depending- on the 

 seepage action of the nearest ditch for drainage. This is by far 

 the safest method, and the number of these pools or salt holes 

 should be taken into account in determiining the distance apart 

 of the ditches in the system. It will be found in some cases 

 where pools are very numerous that the section can be drained 

 more economically l3y putting the main ditches close together 

 and thus cutting down the length of the spur ditches needed 

 than by spacing the main ditches farther apart and thus increas- 

 ing the length of the spurs. 



Various systems for the drainage of the marsh have come 

 into use, the two principal ones being the parallel system and 

 the checkerboard, or circulating system. The parallel system 

 has be'en in use for the most part in New Jersey in recent years, 

 while the best existing example of the circulating system is the 

 one installed on the Jamica Bay marshes in 19 16. Under this 

 system the ditches were run from the upland to the bay or out- 

 let creek at intervals of from 125 to 150 feet apart. At the 

 upland or head end of these ditches a marginal ditch connecting 

 the ends and parallel to the upland was run. At right angles 

 to the first series of ditches, and at distances varying from 175 

 to 200 feet, another series of ditches was run, thus giving the 

 checkerboard effect. 



There are several advantages to be claimed for this system, 

 the principal one being that it is extremely difficult to have the 

 ditches so obstructed by floating sods, etc., that they will not 

 drain at low tide. With so many interesting ditches the water 

 can always run out, although it may be by devious courses. On 

 a recent inspection of the upper end of Jamaica Bay, after some 

 extremely high tides, only one ditch was found which was not 

 draining satisfactorily in a section containing over two million 

 feet. It is therefore obvious that this system will maintain 

 itself with less inspection than the parallel system where a ditch 

 several thousand feet long may become obstructed by a single 

 sod, and therefore fail to drain a large area. It is also evident 

 that this system would seem to be ideal in places where, by 

 reason of lack of maintenance funds or scarcity of labor or 

 inspectors, the system must maintain itself. 



