Effects of feeding by Scirpophaga incertulas larvae

Having argued that S. incertulas infestation in the elongating crop may well not equate with crop loss, the question arises - why not?

As an internal parasite, which enters the rice stem as a first-instar larva and does not leave until it emerges as an adult, S. incertulas depends for its survival on larvae not destroying the stem which they inhabit. The anatomical studies reported above show that in deepwater rice, except in seedlings and in the panicle-bearing internode, larval feeding does not cause significant structural damage or interruption to the flow of nutrients. The passage of a larva through the nodal septum also appears not to have any detrimental effect. This is because the septum is composed solely of pith, i.e. undifferentiated ground tissue; penetration cannot affect nutrient conduction and, the stem being cylindrical, it should not affect the structural strength of the node.
	The exit hole (left) bored prior to pupation by the last instar larva is a potential weak point, but it is small relative to the large stems characteristic of deepwater rice. Entry by bacterial or fungal rot organisms is conceivable, although the hole is sealed with silk. The large elongated stems are fibrous, and it is common for the lower submerged internodes to die without loss of the stem as a connection between the original soil roots and the upper stem.
	Deepwater rice varieties are characterized also by the development of nodal roots (left) which assimilate nutrients from the water and thus counteract this death of the lower stem. In the open flooded fields, uprooting of deepwater rice plants is not uncommon, but it has been shown that yields may still be near that of non-uprooted plants (Alim & Zaman 1958).
Larval feeding in young plants, with commensurately narrow tillers, can cause tiller death (deadhearts), but two natural factors reduce this problem in deepwater rice. First, dry sowing is the norm and the young plants usually get their moisture from residual soil moisture or sparse spring rainfall, whereas S. incertulas prefers wetland conditions and high humidity. Second, being the end of the dry season, S. incertulas adult populations are at the annual minimum. In the mature crop, whiteheads, or unfilled dead panicles, are an obvious direct loss of yield. Again the borer larva is feeding in a narrow stem (right), and serious to total interruption of nutrient conduction is almost inevitable. These findings are contrary to the statements by Catling & Islam (1982b), but the apparent tolerance of S. incertulas by deepwater rice is consistent with the opinion that deepwater rice varieties form a primitive group of cultivated rices. Chowdhury & Zaman (1970) described this group as probably a direct descendant of the most prevalent wild rice species of Bangladesh, Oryza sativa var. fatua. The selection of paddy rice varieties with shorter, narrower stems could have led to inherently greater damage to vascular tissues by larval feeding and, thus, to the formation of deadhearts throughout the growth of the crop.

Correlation between infestation and yield

The evaluation of the correlation between infestation and yield in 1981, using data from agronomy trials (Table I, for example) suggested that pre-flood infestation levels below 24% and infestation levels from harvested stems below 42% had not affected yield. Detailed dissections of 100 panicle-bearing stems showed no difference in panicle weights on infested or uninfested stems. The highest grain sterility and most whiteheads were associated with infestation in the terminal, panicle-bearing internode.

We followed this up in 1982 by sampling all the project trials, and some 25,000 stems were collected for dissection. Taken as a whole, with a range of mean infestation levels (27-60%) and variability in the levels of natural borer infestation (C.V. around 40%), the results showed no consistent correlation between stem infestation and yield (84 separate correlations were calculated, in some cases with d.f.>30).

For instance, in the factorial and variety trials at four sites (Table IV & Table V), only three correlations were significant. For the varieties Ejuli Digha and BR 306-B-3-2 they were positive, indicating higher infestation leading to higher yield, and for Kartiksail the correlation was negative, indicating higher infestation had reduced yield. Results for these three varieties from other sites showed no significant correlations. The combined data for the variety Gilamite, from three transplanted trials, showed a significant negative correlation. The previous year, Gilamite had been heavily infested (Table I) but the yield had been sustained by a compensatory production of high numbers of secondary, or nodal tillers which counterbalanced their lower individual panicle weights.

Two detailed studies of stems bearing panicles showed that it was primarily infestation of the terminal, panicle-bearing internode that led to loss in grain weight (Table VI). This was particularly marked in the variety Chamara where, despite an infestation level of 97% in the 254 stems dissected, no significant grain loss had occurred unless the terminal internode was infested. Dissection of 594 stems of Sada Pankaish (overall infestation 37%) showed infestation in the terminal, third and fourth internodes to affect grain filling. In both varieties, whiteheads associated with infestation in the terminal internode were the predominant cause of grain loss. In this study, the varieties were collected from widely separated locations but samples of Sada Pankaish from fields at the Chamara location showed 41% infestation. The difference may be that Chamara with thicker stems is more susceptible to borer attack but, to compensate, having thicker, tougher stems could enable it to better withstand larval feeding. Catling found that whereas in 1978 thicker stemmed varieties had higher infestations this was not the case in 1979 (Catling 1981).

Finally, the specific study of rice stems with whiteheads showed that, although borer infestation down to the sixth internode may be associated with whiteheads, 94% of the 205 whiteheads were associated with infestation in the terminal internode. In this study, the majority of infestations were by two or more larvae whereas in earlier findings larvae usually were present singly. Presumably, this was due to a late season population explosion as the highest numbers of larvae found in multiple infestations were 23 of S. incertulas and 47 of Sesamia inferens.

Results from West Bengal, India (Datta et al. 1985; Datta et al. 1985), from Vietnam (Van Huynh et al. 1986) and from Thailand (Catling et al. 1987) showed the same pattern of whiteheads as the main source of crop loss. In Vietnam, about 5% whiteheads and 3-4% stem infestation were found in 1984 (data from 150 plants). In West Bengal, there was no yield when the terminal internode was infested and somewhat lower yield when the 2nd to 12th internodes (as a group) were infested (data from 20 plants per group); infestation in 1983 related to high borer incidence in late September and was restricted mostly to the top three internodes. In Thailand, when whiteheads were omitted, the mean panicle weights on infested stems were higher than in uninfested stems. Similar studies in Bangladesh in 1978 (unpublished ODA-BRRI project records) found no relationship between infestation and yield in samples from crop-cut surveys. A panicle weight assessment, with dissections of some 180-200 stems from each of four fields, showed that in three fields panicles from infested stems were lighter by 4.5-12.7%, but in the fourth field such panicles were heavier by 21.7%.

The results cited by Catling et al. (1987) from small-scale pot and cage experiments in Bangladesh did appear to suggest that stem infestation has an effect on yield, but my careful examination of the original records showed that a different interpretation of the results is possible. For instance, the 1978 pot study with Habiganj Aman gave a yield reduction in the infested pots of 9.6 g/pot (36.7%). This could have been accounted for by the initial deadhearts in the infested pots, which at 4.3 per pot (each being direct loss of a primary panicle weighing 2.2 g), amounted to 9.5 g. This initial loss was offset by the development of secondary tillers, but these typically had lighter panicles. The subsequent 20% stem infestation may well have had no effect on the final yield. Also, in the small-scale experiments, the depth of water never exceeded 1 m, thus precluding the normal elongation of the plants which would occur under field conditions of water 2-3 m deep.

continue

href="\dwr\ysb13.htm"