Ian R. Walker

Meticulous preparation, sorting, and identification of midge remains is requisite for analysis of past communities. As experienced chironomid palaeoecologists will attest, these procedures are extremely tedious. A carefully cleaned sample is greatly appreciated when the sorting operation begins.

Small lakes in forested watersheds usually contain sediments with a high organic content. My experience suggests that 1 or 2 ml of wet sediment from such lakes provides 50 to 100 head capsules, sufficient for analysis (Walker and Paterson, 1983; Walker and Mathewes, 1987). Inorganic sediments, such as those frequently encountered in late glacial deposits, may yield much lower concentrations.

However, head capsule concentrations vary greatly among sites. Deevey (1942) has reported maximum concentrations of 68 head capsules/ml at Linsley Pond, but according to Frey (1964), Deevey (1955) later encountered 1700/ml at Pyramid Valley, New Zealand. The highest concentration yet reported, nearly 8000/ ml, was tallied for Eight Lake, Alaska (Livingstone et al., 1958). Stahl (1959) reported concentrations of 6 to 651 ml. Sediments of the Schöhsee yield concentrations ranging from 0 to 260/ml (Hofmann, 1971). Warwick (1980), for the Bay of Quinte, Lake Ontario, describes concentrations varying from 4.6/ml at the sediment surface to 124/ml at the 1.14 m depth. The range observed in Warwick's (1980) study may result from several factors including compaction of the deeper sediments, recent increased sedimentation rates and changes in head capsule production.

Samples are usually deflocculated in 5 to 10% KOH prior to analysis. In calcareous sediments. an acid wash (10% HCI) may also be necessary. Warwick (1980) emphasizes the delicate nature of midge remains and the need for mild treatments. He recommends the use of 8% KOH at 60°C for 30 minutes. Apparently chitinous structures may be bleached and deformed by high temperatures (Warwick, 1980). It is important to remember that although chitinous structures are very resistant, exuviae of early instars, if present, may contain little or no chitin (Iovino, 1975). Also, harsh treatment may disarticulate head capsule features.

Following the above chemical treatments, head capsules are generally separated from finer debris by means of sieving. Methodology varies, but a 100 µm or finer mesh sieve will retain most head capsules and is recommended (Walker and Paterson, 1985). Subsequently, the residue is back-washed from the sieve into a container and stored wet until the material is sorted. Unless sorted shortly after sieving, preservation in 99% ethanol (Warwick, 1980) is advisable.
A Bogorov counting tray

Sorting of wet residues may be accomplished in a petri dish, watch glass, or Stender dish at magnifications of 25 to 50 x (Warwick, 1980). To limit the possibility of overlooking remains, I prefer to sort head capsules from sediments at 40 to 50 x in a Bogorov counting tray (Gannon, 1971). The particular Bogorov tray which I employ includes 7 parallel grooves cut in a perspex plate. The bottom width of each groove, 4.5 mm, corresponds to the field of view, at 50X, of a Wild M5 stereomicroscope. As remains are located, each is transferred with forceps to coverslips for mounting and identification.

Researchers should note that head capsules of some taxa, especially the Orthocladiinae, readily split into two equal, identifiable halves. Thus, it is important to count these fragments as 1/2 head capsule. Head capsules bearing more than one half of the mentum may conveniently be counted as whole head capsules, and head capsules consisting of less than one half of the mentum may be ignored. This splitting of the head capsules creates difficulties in the application of conventional statistics. If the halves are sedimented independently, perhaps each half should be treated as if representing an individual for statistical purposes. If the head capsules split within the sediment, or during processing, and the head capsule halves remain in close proximity, they would be better treated as halves.

Presentation of results varies among authors. Saw-toothed figures similar to those of pollen diagrams are most common, yet histograms convey the results equally well. In any circumstance, figures convey information more readily than tables of data. Computer programs for plotting pollen diagrams are easily adapted for use with chironomid data.

More critical is the decision to present data as either percentages, counts or influx estimates. Percentages are limited chiefly because changes in the relative abundance of one taxon may result either from an actual change in its influx or a change in the influx of one or more other taxa.

Count data are difficult for the reader to assess since the counts at adjacent levels may not readily be compared unless sedimentation rates remain constant. Readers are required to perform the mental gymnastics necessary to standardize the data. Interpretation of count data suffers from variations in sedimentation rates and sample volume. Consequently, percentage results are usually more convenient.

Ideally, influx rates for each taxon should convey the most information. In practice, influx data are more informative only if accurate measures of sedimentation rate are available. Influx calculations frequently assume a constant sedimentation rate over a broad sampling interval, yet it is likely that sedimentation rates vary considerably in the short term. In calculating sedimentation rates for surface cores, investigators must carefully consider the effect of compaction on apparent rates. Sediment focusing may have important effects (Davis et al., l984), especially during the early history of a lake. Periods of rapid natural or anthropogenic environmental change may induce abrupt changes in sedimentation. Periodic catastrophic events such as forest fires and debris slides in the catchment may induce brief but rapid episodes of sediment deposition. Inaccurate estimates of sedimentation rates cause influx data to reflect these inaccuracies rather than the varying abundance of chironomid taxa.

It is also possible, particularly in small lakes, that as the lake shallows the changing sedimentary environment may alter the influx of chironomid remains without a change in a taxon's actual abundance. A core might record a transition from a profundal stage to a littoral environment. If chironomid remains tend to become concentrated in the sublittoral, then peak concentrations in the middle of a core could reflect this artifact (Walker and Mathewes, 1987).

Authors should cautiously consider how the data are to be displayed and how the changes can most accurately and honestly be portrayed. Perhaps, in future, another form of presentation will be attempted, providing rates of chironomid production. A simple cubic relationship could be calculated between the width of a taxon's head capsule and biomass of the living larva. More perplexing is the task of quantifying each of the taphonomic processes regulating which head capsules are deposited at a site, and which would subsequently be preserved.

For a more detailed account of methods for chironomid palaeoecological research, see Walker (2001).


Davis, M.B., Moeller, R.E., and Ford, J. 1984. Sediment focusing and pollen influx. In: Lake Sediments and Environmental History. Edited by E.Y. Haworth, and J.W.G. Land.

Deevey, E.S., Jr. 1942. Studies on Connecticut lake sediments. III. The biostratonomy of Linsley Pond. American Journal of Science 240: 233-264, 313-324.

Deevey, E.S., Jr. 1955. Paleolimnology of the Upper Swamp Deposit, Pyramid Valley. Records of the Canterbury Museum (N.Z.) 6: 291-344.

Frey, D.G. 1964. Remains of animals in Quaternary lake and bog sediments and their interpretation. Ergebnisse der Limnologie 2: 1-114.

Gannon, J.E. 1971. Two counting cells for the enumeration of zooplankton micro-crustacea. Transactions of American Microscopical Society 90: 486-490.

Hofmann, W. 1971. Die postglaziale Entwicklung der Chironomiden- und Chaoborus-Fauna (Dipt.) des Schöhsees. Archiv für Hydrobiologie, Supplement 40 (1/2): 1-74 (English translation : Fisheries Research Board of Canada, Translation Service No. 2177).

Iovino, A.J. 1975. Extant Chironomid Larval Populations and the Representativeness and Nature of their Remains in Lake Sediments. Ph. D. thesis, Indiana University, Bloomington.

Livingstone, D.A., Bryan, K., Jr., and Leahy, R.G. 1958. Effects of an arctic environment on the origin and development of freshwater lakes. Limnology and Oceanography 3: 192-214.

Stahl, J.B. 1959. The developmental history of the chironomid and Chaoborus faunas of Myers Lake. Investigations of Indiana Lakes and Streams 5: 47-102.

Walker, I.R. 2001. Midges: Chironomidae and related Diptera. In: Tracking Environmental Change Using Lake Sediments. Volume 4: Zoological Indicators, Developments in Paleoenvironmental Research. Edited by J.P. Smol, H.J.B. Birks, and W.M. Last. Kluwer, Dordrecht. pp. 43-66.

Walker, I.R., and Mathewes, R.W. 1987. Chironomidae (Diptera) and postglacial climate at Marion Lake, British Columbia, Canada. Quaternary Research 27: 89-102.

Walker, I.R., and Paterson, C.G. 1983. Post-glacial chironomid succession in two small, humic lakes in the New Brunswick - Nova Scotia (Canada) border area. Freshwater Invertebrate Biology 2: 61-73.

Walker, I.R., and Paterson, C.G. 1985. Efficient separation of subfossil Chironomidae from lake sediments. Hydrobiologia 122: 189-192.

Warwick, W.F. 1980. Palaeolimnology of the Bay of Quinte, Lake Ontario: 2800 years of cultural influence. Canadian Bulletin of Fisheries and Aquatic Sciences 206: 1-118.

©2007 Ian R. Walker. ALL RIGHTS RESERVED

Last Update: 

Recommended Citation