The Pa-tuna for a Large River.
This pa-tuna is always built with the top end on the crest of a swift rapid, and consists of two parallel fences with cross-returns of a single post facing each other at the foot to hold the hinaki. They are exceedingly well built, and very strong considering they are erected in the middle of swift waters from canoes that have to be held in position by poles, and also where the river-bed is composed of boulders and large stones. I am informed by the natives that the fence on the western side is always the shorter, but no reason is obtainable why this is so. Reference to the illustrations will enable the reader to see that this form of weir, composed of two straightened parallel fences, differs widely from the V-shaped weir employed in many rivers, and also from the lamprey-weir, which extends from the river-bank outwards at a right angle to the current. (See Plate XXV, figs. 1 and 2.)
After getting all the poles, timbers, and lashings together, it takes from four to six men at least seven days' hard work to construct the simplest form of this pa-tuna. The hardwood stakes of kopuka* are, as a rule, about 4 in. m diameter, and they are driven into the heavy shingle from 2 ft. to 2 ft. 6 in. with a sort of wooden maul, called a ta.
The weir is, as a rule, from 50 ft. to 60 ft. long and about 20 ft. wide, and the work is commenced at the crest of the rapid and continued downstream. After a number of poles have been driven in, two horizontal timbers are lashed on, one below the other, after which more stakes are driven, it being easier to keep in line with guiding-timbers on top. A long and very heavy totara log, from 12 in. to 18 in. in diameter, is then lashed to the stakes at about low-water level, and further held in position by another row of stakes driven at an angle, the top of the stake finishing flush with the inside of the fence (fig. 6). The last post down-stream is clear of the heavy log, and only held by the top horizontal timber, so as not to interfere with the poha sliding up and down. This will be seen in the picture of the pa-tuna on Te Aute-mutu Rapid (see Plate XXVI).
Considerable judgment was required in setting the fences at the proper angle against the current, and because of care in this matter, combined with good workmanship and position, some pa-tuna took more fish than others. The angle of fences was of the utmost importance, and always they ran into the current to a greater or less degree according to the arrangement of the stakes. If parallel with the current, or nearly so, few fish were intercepted, and if at too great an angle the eels escaped through the fence.
In a close arrangement of stakes, as the pa-tuna at Kauwae-roa (Plate XXV, figs. 1 and 2) a greater angle is given than in the pa at Te Aute-mutu (Plate XXVI), where the stakes are wider apart. The double fence was only for the purpose of intercepting more fish.
A bad architect superintending the construction of a pa-tuna was the object of much derision, and his failure was known throughout the district. An unsuccessful pa was always pulled down.
It is said that when eels travel up-stream they usually take the deepest and darkest water, taking advantage of every help, while lampreys keep close to the edge, especially in swift water. The log with its double row of stakes causes a sort of backwater right up the full length of the weir, and provides an easy passage for the wily tuna, which he is not slow to take advantage of. At the top of the pa-tuna a sloping rounded log, carefully smoothed, is fixed so as to turn the eels and cause them to be thrown back by the current, which carries them down into the poha before they regain shelter. At the foot of the weir two posts are driven in about 5 ft. away from the fences, one on either side, facing each other, and strongly braced to the main structure, their object being to hold the, poha frame.
[Footnote] * I believe the name kopuka is peculiar to the Whanganui River natives, the names kanuka and maru being used for the wood (Leptospermum ericoides) in other places.
Usually there are sliding logs that work between these posts and the fences, held by the force of water, and also a rope that lifts or lowers them, together with the poha frame of the inverted Y pattern (see later), which is fixed to the hinaki in the manner before described (see fig. 7). These angles, being right across the current, are soon broken by driftwood, and, as a rule, have to be renewed or repaired annually. All the lashings used in the construction of the pa are of aka or kareao vines; no pegs or nails are ever used even in modern times. Usually the fences are lowest at the top of a rapid, gradually rising as they go down-stream.
Fig. 7.—Attachment of the poha. 1, angle brace; 2, sliding timber used to raise and lower the poha, 3, hinaki; 4, ropes of twisted kareao.
The names of the parts of the pa-tuna are as follows: The stakes are called pou; the top horizontal timber, uaua, sometimes (I think, correctly) huahua; the second horizontal timber, kaiwai; the heavy bottom totara log, huahua-kaiwai; the angle stakes holding same, noko; the angle log at head of weir, noko-panawai; the return angle or wing at foot, hoi; the side posts holding same, pou-riri (sometimes turu); the sliding timber, rango; the same timber when fastened down, huapae; the water between the fences, ihonui; water outside of fences, auroa.
When a fresh is in evidence two baskets are placed in position on each side, one above the other, as in the case of the small pa-tuna. Of course, in this particular style of weir a great many eels must pass without being caught; but it would be quite impossible to net a large river in this manner thoroughly, owing to the logs and debris coming down. No doubt if the fences were to converge gradually they would be more effective in fishing, but they would be more liable to be destroyed, as the drift timbers would be caught and the weight of waters would soon be irresistible. However, a very large number of fish are taken, usually in April. The only time I saw this pa-tuna being worked upwards of half a ton of fish was taken out within twenty-four hours. This was during the tuna-heke migration.
A very large pa-tuna capable of holding eight or more hinaki is called pa-tuna waharoa. There is also another built on a zigzag principle, but neither of these have I seen, nor have I been able to obtain any description of them.