Survival and Vertical Distribution of Larvae of Culicoides variipennis (Diptera: Ceratopogonidae) in Drying Mud Habitats BRADLEY A. MULLENS AND JOHN L. R O D R I G U E Z Department of Entomology, University of California, Riverside, California 92521

KEY WORDS Insecta, Culicoides variipennis, control, habitat THE PRIMARY VECTOR of bluetongue virus in

North America is the biting midge Culicoides variipennis (Coquillett) (Jones et al. 1981). This midge exploits manure-polluted or alkalinesaline aquatic habitats for immature development, and larvae most commonly are found in surface mud near the water's edge (Barnard & Jones 1980). Many of these habitats (e.g., rainwater pools, water trough overflow, or evaporation beds for dairy wastewater) are ephemeral. Such habitats may be colonized quickly by C. variipennis after flooding and can produce large numbers of adults (Mullens & Rodriguez 1988, Mullens 1989). However, larvae probably are stranded frequently in mud without free water because of evaporation, pumping for irrigation, etc. Observations and prior experimental water level manipulations (Mullens & Rodriguez 1989) indicated that larvae could persist for several days to perhaps weeks in mud without free water. There was also some evidence that larvae either moved deeper into mud to withstand drying conditions or that larvae in deeper mud were better able to survive (Mullens & Rodriguez 1989). The ability of C. variipennis larvae to survive unfavorable periods, and perhaps resume development after reflooding, has considerable implications for control through water management. The present study was designed to examine the ability of C. variipennis to survive in drying mud habitats under simulated field conditions. Materials and Methods

Habitat mud was obtained from the surface 15—20 cm of a dairy wastewater site in Mira

Loma, Calif., utilized by C. variipennis. The mud was dug from 1 to 2 m back from the waterline when the water level had been low for several weeks. This substrate was autoclaved to kill existing organisms and thoroughly homogenized in a cement mixer. The same soil was stored and used for all experimental trials. An aliquot of the soil was submitted to the Division of Agricultural and Natural Resources Soil Diagnostic Laboratory, Cooperative Extension, University of California, Riverside, for determination of physical characteristics and organic matter content. Before each experimental trial, water from the dairy wastewater pond mentioned above was filtered through a 240-mesh sieve to exclude arthropods but allow passage of microorganisms important for C. variipennis larval nutrition (Jones et al. 1969). This water was mixed with distilled water (1:9 ratio), and added to the soil until the mud flowed easily when poured. Plastic cylinders (Georgis & Poinar 1983) (10 cm long, 7.5 cm diameter) were cut into sections. The top section was 2 cm, followed by sections of 1, 1, 2, 2, and 3 cm (bottom section). A solid plastic cap sealed the bottom, and the sections were joined with masking tape. Mud was added to each tube the day before the experiment began. Tubes were filled to 1 cm below the top rim, and were left open for the duration of the experiment. This arrangement allowed later sectioning of the mud cylinder at depths of 0-1,1-2,2-3, 3-5, 5-7, and 7-10 cm. Trials were done in September and October of 1989 and 1990, when the numbers of C. variipennis in the region were high (Mullens & Lii 1987).

0022-2585/92/0745-0749$02.00/0 © 1992 Entomological Society of America

Downloaded from http://jme.oxfordjournals.org/ by guest on June 7, 2016

J. Med. Entomol. 29(5): 745-749 (1992) ABSTRACT The ability of third and fourth instars of Culicoides variipennis (Coquillett) to survive in drying sandy loam soil was tested under simulated field conditions. Two hundred larvae were added to each of a number of soil-filled, plastic tubes, which were buried in the field and retrieved after 2, 5, and 7 d. Of 306 pupae or pupal exuviae recovered, 98.1% were in the top 2 cm of mud. Estimated larval mortality on day two ranged from 25.6 to 87.1% among three trials, and 92.3% of live larvae were in the top 2 cm. Estimated larval mortality on day 5 ranged from 81.4 to 97.2%, and 75.5% of live larvae were in the top 2 cm. On day 7 mortality was 95.2-100%, and 63% of surviving larvae were in the top 2 cm. Some larvae dispersed as deep as 7—10 cm (the maximum depth in the tests). Larval tolerance of unfavorable drying conditions may allow relic populations to persist in some situations, but probably this is of little concern in control of this species through water management in most settings.

746

Vol. 29, no. 5

JOURNAL OF MEDICAL ENTOMOLOGY

Pupae (n= 306)

100252

80-

S

601

o

"5

4048

20-

0-1

1 - 2

2 - 3

3 - 5

5 - 7 7 - 1 0

Depth (cm) Fig. 1. Distribution of C. variipennis pupae and pupal exuviae according to depth in the substrate. Actual numbers recovered are noted above histogram bars.

was used as the independent variable, and the percentage of total larvae in that tube per depth stratum was used as the dependent variable. Slopes of the regression lines were compared by an approximate t test (not assuming equal variances) to determine whether the depth distribution of surviving larvae differed among trials or days (Snedecor & Cochran 1980). All analysis was done using the Minitab statistical package (Ryan et al. 1985). Results

The soil used was a sandy loam with 2% organic matter and 72% sand, 20% silt, and 8% clay. Water drained readily through this soil, and larvae were without free water within minutes after they were added to the tubes. Larvae could be observed crawling on the mud surface for several minutes after the free water drained away, but only two larvae actually were seen to crawl up the lip of the tube and onto the surrounding soil. The remaining larvae quickly burrowed into the mud surface. Significant pupation occurred only in trial three (12.8% of larvae added), where estimates of larval mortality were corrected for pupation. A total of eight pupae, pupal exuviae, or both were recovered in trials 1 and 2, so estimates of larval mortality did not require correction. Of the 306 pupae or pupal exuviae recovered, 82.4% were in the top 1 cm (98.1% in the top 2 cm), though one pupa was recovered at the 5-7 cm depth (Fig. 1). This agrees with prior work that demonstrates the surface distribution of C. variipennis pupae (Barnard & Jones 1980). Dead larvae were detected, but usually were difficult to see because they dried and shriveled

Downloaded from http://jme.oxfordjournals.org/ by guest on June 7, 2016

Surface mud (upper 1-2 cm) was removed from the edges of a nearby Mira Loma dairy wastewater pond habitat the day before each trial. Late-instar larvae were extracted from the mud by washing through 40- and 100-mesh sieves. Very large larvae (near pupation) tended to be retained on the 40-mesh sieve. Larvae retained on the 100-mesh sieve were predominantly third and fourth instars and were backwashed into white enamel pans. Larvae were removed with pipettes into groups of «=200. On the morning of day zero, 15 tubes filled with the wet soil were transported to the Agricultural Experiment Station near the University of California, Riverside campus. They were buried in unshaded, wet mud until the mud surface in each tube was level with the surrounding mud surface. One group of 200 C. variipennis larvae was added in ==10 ml of distilled water to the soil surface of each of 12 tubes. The remaining three tubes received water only. The mud surface in the tubes was watched for ^30 min after the water and larvae were added, to observe whether larvae would crawl out of the tubes or burrow into the mud within the tubes as the water receded. On the morning of days 2, 5, and 7, four tubes to which larvae had been added and one tube to which water only had been added were excavated and returned intact to the laboratory. The masking tape was removed from the outside of the tubes. A thin sheet of stainless steel could then be inserted between the tube sections from the side to separate the depth strata cleanly into separate cups. Aliquots (30-50 g) of each mud layer from the tube without larvae were placed into weighing dishes, immediately weighed (±0.1 g), oven-dried for 2 d at 50°C, and reweighed to obtain estimates of the water content. Mud layers from the remaining tubes were individually washed through 20- and 100-mesh sieves. The 20-mesh sieve retained coarse debris, which was backwashed into a white enamel pan and examined for 1-3 min for C. variipennis larvae, pupae, or pupal exuvia. Debris on the 100-mesh sieve was subjected to flotation with saturated MgSO4 as described by Mullens & Rodriguez (1984), and C. variipennis immatures were removed with a pipette and counted. The solution was stirred at 2-min intervals and examined continuously (under a 3x circular lens with light) until no larvae, pupae, or pupal exuvia were seen for two consecutive intervals beyond the initial 2-min examination period. The entire experiment was repeated three times. Data on total numbers of larvae per tube (depths pooled) were analyzedfirstby analysis of variance (ANOVA) to determine whether an interaction between trials and sample days existed. Individual trial-day combinations that yielded at least 20 larvae per tube then were subjected to regression analysis. Depth (top four layers only)

September 1992

MULLENS

-

40-1

—«—

1

Air Temp

6"

-A

< ^

30-

|

0— 1Max - - © - • 1Min

©-*

20;

•- -•—

•. - • •

10

* 0

o

o— —•a"'

1

2

..o-

3

°-.

-A

—•—

2Max

—-•—

2Min

A— 3Max —-A— 3Min

"•0

4

5

6

747

lODRIGUI

7

Day

100n 80 i

2RH 3RH

Fig. 2. Maximum and minimum ambient air temperature and average relative humidity (RH) levels during the three experimental trials.

quickly. Numbers of live larvae recovered are presented in Table 1 by trial, sampling day, and depth. Assuming no larval escape from the tubes and a high level of late-instar extraction efficiency (Mullens & Rodriguez 1984), estimated larval mortality on day 2 was 87.1% in trial 1, 25.6% in trial 2, and 57.0% in trial 3. By day 5, mortality was 97.2% in trial 1, 98.4% in trial 2, and 81.4% in trial 3. Nearly all larvae were dead

Table 1. Numbers of live C. variipennis larvae (x ± SE) recovered (of 200 added per replication) and moisture at selected soil depths after exposure to field drying conditions

Day 2

5

7

Soil depth, cm 0- 1 1- 2 2- 3 3- 5 5- 7 7-10 0- 1 1- 2 2- 3 3- 5 5- 7 7-10 0- 1 1- 2 2- 3 3- 5 5- 7 7-10

Trial 1 No. larvae %H20 23.0 2.8 0.0 0.0 0.0 0.0 0.5 1.8 1.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

2.4 0.9 0.0 0.0 0.0 0.0 0.5 0.6 1.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

32 32 29 30 30

30 31 23 24 25 26 25 6 19 22 24 27 26

Trial 2 No. larvae 122.8 dt21.6 12.3 dt 2.3 10.3 dt 5.6 2.0 dt 1.4 1.0 dt 0.4 0.5 dt 0.5 0.0 dt 0.0 0.0 dt 0.0 0.3 dt 0.3 2.3 dt 1.6 0.3 dt 0.3 0.3 dt 0.3 0.0 dt 0.0 0.0 dt 0.0 0.0 dt 0.0 0.8 dt 0.5 0.0 dt 0.0 0.0 db 0.0

%H 2 O 45 39 33 32 32

33 7 20

23 27 28 28 3 12 19

25 25 26

Trial 3 No. larvae 59.5 dt 11.8 dt 2.8 dt 1.3 dt 1.3 dt 0.0 dt 17.0 dt 10.3 dt 2.3 dt 2.8 dt 0.0 dt 0.0 dt 3.8 dt 2.0 dt 1.8 dt 0.8 dt 0.0 dt 0.0 dt

11.1 2.8 0.9 0.5 0.9 0.0 2.6 1.1 1.7 2.1 0.0 0.0 2.8 0.8 0.9 0.5 0.0 0.0

%H 2 O 23 22 23 23 23 26 26 26 24 23 21 24 13 19 18 18 21 19

Downloaded from http://jme.oxfordjournals.org/ by guest on June 7, 2016

1RH

by day 7, with estimated mortality of 100, 99.6, and 95.2% for trials 1—3 respectively. The larval mortality did not necessarily correspond well with the temperature (and mud dryness) in the different experimental trials. For example, larval mortality was highest in trial 1, despite relatively low temperatures (Fig. 2). Temperatures were highest, and humidity lowest, in trial 2, but mud moisture level and larval recovery stayed high through day 2 (Table 1). Continuing high temperatures (Fig. 2) and dry surface mud conditions on days 5 and 7 (Table 1) presumably contributed to low numbers of larvae recovered later in trial 2 (Table 1). An initial ANOVA (main effects were trial and day) on the total number of larvae recovered per tube indicated that the trial x day interaction was highly significant (F = 14.63; df = 4, 27; P < 0.001). This was influenced by variability in larval survival among trials (Table 1). Surviving larvae generally remained in the top 2 cm, although a few migrated downward as deep as 7-10 cm as early as day 2 (Table 1). Extremely low numbers of larvae recovered on day 5 in trials 1 and 2 and on day 7 in all trials made statistical analysis inadvisable. Although variable among trials, there was some indication that larvae surviving to days 5 and 7 were found in deeper mud strata than on day 2 (Table 1). On day 5, 84.2% of live larvae were in the top 2 cm in trial 3 and 63.8% in trial 1. In contrast, there were no live larvae in the surface 2 cm on day 5 in trial 2 (where surface mud moisture was very low), and the 12 larvae recovered all were 2—10 cm deep (Table 1). By day 7, the three surviving larvae in trial 2 were 3—5 cm deep, whereas 78.3% of larvae in trial 3 were still in the surface 2 cm. Data for day 2 (all trials) and day 5 (trial 3 only) were selected for further analysis, based on the

748

JOURNAL OF MEDICAL ENTOMOLOGY

Discussion Previous water level fluctuation experiments demonstrated that C. variipennis larvae remained in place after a rapid water level drop, and larval mortality apparently was high by 7 d after the water receded (Mullens & Rodriguez 1989). A confounding factor in those experiments was the possibility of larval movement out of the fairly narrow sampling zone. Larval mortality estimates from our earlier studies (Fig. 3 in Mullens & Rodriguez [1989]), however, were comparable with estimates seen in the present study, suggesting that horizontal larval movement was not a significant factor in that study. The present experiments confirmed that most larvae will perish within 7 d after they are stranded in mud without free water, unless they are mature enough to pupate and emerge. Earlier experiments also indicated that some larvae may move deeper into the substrate over time, and a few larvae were found alive up to 19 d after a drop in the water level (Mullens & Rodriguez 1989). The present experiment also showed that the few C. variipennis larvae surviving on days 5 and 7 tended to be deeper, with some as deep as 7-10 cm. The sandy nature of the habitat mud used tended to minimize cracking as the mud dried, although the soil did draw away from the sides of the tubes slightly in some replicates. The fissures may have provided the larvae with an easier route to retreat to deeper strata, as probably occurs naturally in the field. Live C. variipennis larvae recovered from the drying mud often were very sluggish, but sometimes would become more active after immersion in water. However, we did not see evidence of an ability to desiccate and remain in a viable inactive state for extended periods, as is known

for the chironomid midge Polypedilum vanderplanki (Hinton) in African rain pools (Hinton 1968). P. vanderplanki essentially can use an ephemeral resource as a permanent one, with characteristics such as body size being much less variable than those of midges sharing the habitat (i.e., Chironomus imicola Kieffer) that lack this ability (McLachlan 1983). Dead larvae of C. variipennis were desiccated totally and did not revive in water. Larval survival of C. variipennis might be better in mud under cooler, moister conditions; for example, clay soils would retain moisture for relatively longer periods of time than the sandy loam used in our experiments. Clay soils also could provide more and deeper cracks, which could serve as routes of access for the larvae. Although the ability of C. variipennis larvae to endure short-term, unfavorable habitat conditions might be quite important in maintaining relic populations, it probably has little significance for water management to suppress this species in most situations. Acknowledgment We appreciate the critical review of an earlier version of this manuscript by F. Holbrook, M. Robertson, and W. Tabachnick (USDA-ARS, Laramie, Wyo.) and thank Coralie Dada for technical assistance. We thank an anonymous reviewer and C. Huszar (Department of Statistics, University of California, Riverside) for their advice on the analysis.

References Cited Barnard, D. R. & R. H. Jones. 1980. Culicoides variipennis: seasonal abundance, overwintering, and voltinism in northeastern Colorado. Environ. Entomol. 9: 446-451. Georgis, R. & G. O. Poinar, Jr. 1983. Effect of soil texture on the distribution and infectivity of Neoaplectana carpocapsae (Nematoda: Steinernematidae). J. Nematol. 15: 308-311. Hinton, H. E. 1968. Reversible suspension of metabolism and the origin of life. Proc. R. Soc. (B) 171: 43-47. Jones, R. H., A. J. Luedke, T. E. Walton & H. E. Metcalf. 1981. Bluetongue in the United States, an entomological perspective toward control. World Anim. Rev. 38: 2-8. Jones, R. H., H. W. Potter & S. K. Baker. 1969. An improved larval medium for colonized Culicoides variipennis. J. Econ. Entomol. 62: 1483-1486. McLachlan, A. 1983. Life history tactics of rain pool dwellers. J. Anim. Ecol. 52: 545-561. Mullens, B. A. 1989. A quantitative survey of Culicoides variipennis (Diptera: Ceratopogonidae) in dairy wastewater ponds in southern California. J. Med. Entomol. 26: 559-565. Mullens, B. A. & K.-S. Lii. 1987. Larval population dynamics of Culicoides variipennis (Diptera: Ceratopogonidae) in southern California. J. Med. Entomol. 24: 566-574. Mullens, B. A. & J. L. Rodriguez. 1984. Efficiency

Downloaded from http://jme.oxfordjournals.org/ by guest on June 7, 2016

recovery of ^20 larvae in each tube. Because absolute numbers per tube were similar within a trial—day combination but differed greatly among trials, the number of larvae per depth layer was converted to a percentage of the total for that tube and day. After a log-log transformation (n + 1), the percentages were plotted against depth, excluding the bottom two layers, which generally held no larvae. The slope of each linear regression reflected larval distribution among depths. On day 2, larvae were similarly distributed in trials 2 and 3 (t = 0.51, df = 24, P > 0.05), but were more restricted to the surface in trial 1 (t = 5.54 with df = 30 versus trial 2 and t = 7.16 with df = 17 versus trial 3, P < 0.001). Sufficient numbers of larvae for analysis on day 5 were recovered only in trial 3; based on regression slopes, larvae in this trial were distributed at greater depths when compared with each of the day 2 trials (t = 8.44 with df = 28 versus trial 1, t = 4.02 with df = 27 versus trial 2, and t = 4.18 with df = 19 versus trial 3, P < 0.001).

Vol. 29, no. 5

September 1992

MULLENS

& RODRIGUEZ:

of salt flotation for extraction of immature Culicoides variipennis (Ceratopogonidae) from mud substrates. Mosq. News 44: 207-211. 1988. Colonization and response of Culicoides variipennis (Diptera: Ceratopogonidae) to pollution levels in experimental dairy wastewater ponds. J. Med. Entomol. 25: 441-451. 1989. Response of Culicoides variipennis (Diptera: Ceratopogonidae) to water level fluctuations in ex-

SURVIVAL OF

Culicoides

IN MUD

749

perimental dairy wastewater ponds. J. Med. Entomol. 26: 566-572. Ryan, B. F., B. L. Joiner & T. A. Ryan, Jr. 1985. Minitab Handbook, 2nd ed. Duxbury, Boston. Snedecor, G. W. & W. D. Cochran. 1980. Statistical Methods, 7th ed. Iowa State University Press, Ames. Received for publication 18 April 1991; accepted 18 February 1992.

Downloaded from http://jme.oxfordjournals.org/ by guest on June 7, 2016

Survival and vertical distribution of larvae of Culicoides variipennis (Diptera: Ceratopogonidae) in drying mud habitats.

The ability of third and fourth instars of Culicoides variipennis (Coquillett) to survive in drying sandy loam soil was tested under simulated field c...
501KB Sizes 0 Downloads 0 Views