Next Article in Journal
A Stranded Mola alexandrini (Ranzani, 1839) from the North-Western Adriatic Sea: Integrating Morphological and Molecular Evidence with a Preliminary Morphometric Framework
Previous Article in Journal
Bioaccumulation and Health Risk Assessment of Some Metals in Common Carp—A Lake Perspective
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Hydrobiology
Volume 5
Issue 3
10.3390/hydrobiology5030022
hydrobiology-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Effect of Litter Type on Colonization and Desertion of Food Patches by a Litter-Shredding, Limnephilid Caddisfly

by
Barry R. Taylor
Department of Biology, St. Francis Xavier University, Angitonish, NS B2G 2W5, Canada
Hydrobiology 2026, 5(3), 22; https://doi.org/10.3390/hydrobiology5030022
Submission received: 30 April 2026 / Revised: 5 July 2026 / Accepted: 7 July 2026 / Published: 9 July 2026
Browse Figures
Versions Notes

Abstract

Larvae of Pycnopscyche guttifer, a large, leaf-shredding caddisfly, have become common in rivers of northern Nova Scotia, Canada, where they may greatly accelerate decomposition rates of leaf litter. How the larvae find litter patches in the complex environment of a stream bottom, and how they decide which litter types to feed on, have not been fully examined. In field and laboratory experiments, I determined the preference of P. guttifer to conditioned leaf litter from four species of deciduous trees ranging from recalcitrant and nutrient poor (red oak) to soft and nutrient rich (speckled alder). In field experiments, larvae exhibited a strong preference toward speckled alder and red maple litter, while trembling aspen and especially red oak litter were rarely used. The abundance of larvae on red maple litter varied according to whether aspen or oak litter was nearby. In laboratory choice experiments, larvae searched actively, moving in straight lines in random directions. They deserted less-favoured litter types sooner than favoured litter types, but sometimes abandoned favoured litter types as well. In a complex environment containing many kinds of litter in scattered patches, a restless larva that periodically searches for new food patches may maximize its intake of energy or essential nutrients and obtain a selective advantage over those that remain on acceptable patches.

1. Introduction

In temperate and boreal forest streams, leaf litter from the bankside vegetation is the primary source of reduced carbon and essential nutrients that powers the aquatic ecosystem [1,2,3]. Leaf litter falling into a stream decomposes according to a complex process involving physical forces (leaching, fragmentation), microbial metabolism (particularly by aquatic fungi), and consumption by litter-feeding invertebrates (shredders) [4,5,6]. In addition to direct consumption, shredders accelerate mass loss from decomposing litter by tearing off fragments of the decomposing leaf and thereby reducing particle size (see reviews in [7,8]).
The primary sources of nutrition for leaf-shredding invertebrates appears to be both the leaf tissue itself and the micro-organisms, chiefly fungi, growing on it [7,8,9]. The microbes improve the palatability and nutritional quality of the leaf tissue by increasing the nitrogen content and partially breaking down fibrous tissues [10]. In feeding experiments, most shredders show clear preferences for leaf litter “conditioned” by several weeks of tissue softening and fungal colonization over freshly fallen litter [11,12,13,14], although there are exceptions [15]. Shredders tend to prefer leaf litter from some tree species, especially Alnus spp., over others [16,17,18,19]. Experiments with the limnephilid caddisflies Hesperophylax sp. and Psychoglypha sp. showed that larvae fed selectively on fungi-colonized patches within individual leaves, and preferences varied according to fungal species and colonization time [20]. Pycnopsyche guttifer, the limnephilid caddisfly studied in the current work, showed consistent preferences for birch leaves colonized by some species of fungi over others [21]. On the other hand, larvae of Pynopsyche gentilis showed little selectivity among fungal species on oak leaves, although they always preferred colonized leaves [9].
In streams and shallow rivers of northern Nova Scotia, Canada, Porter and Taylor [22] reported a recent increase in densities of the large, leaf-shredding caddisfly Pycnopsyche guttifer which greatly accelerated mass loss from decaying leaf litter in spring. Mass loss in the presence of P. guttifer followed an approximately linear pattern equivalent to 3.0–5.4% d−1, which continued until the litter sample was entirely consumed. The appearance of P. guttifer in large numbers suggests a shift in leaf litter processing from microbially mediated to shredder dominated in these water courses.
Given the profound effect of P. guttifer on mass loss rates, the feeding preferences of this insect may be pivotal to how long various types of leaf litter persist in the stream, which in turn affects the total supply of litter at a time of the year when organic material may be in short supply [23,24]. In autumn, especially after leaf fall, most leaf litter probably accumulates in mixed-species packs. In spring, when litter is scarce, larvae would likely encounter individual leaves unburied by spates or blown in from the forest floor.
Porter and Taylor [22] noted that P. guttifer larvae rarely colonized red oak leaf litter, of which the decomposition rate was largely unaffected, but readily colonized red maple and speckled alder leaves. In experiments with larvae confined in litter bags, the consumption rate on alder leaves (9–11% wk−1 for two larvae) was noticeably greater than that on maple leaves (5–7% wk−1). It appears, therefore, that P. guttifer larvae, congruent with other shredders, prefer some litter types over others. These preferences are presumably expressed in movements of larvae on the stream bottom in search of litter patches, coupled with decisions of whether to remain on a newly discovered patch or move on to another.
Hence, the possibility that P. guttifer prefers to feed on particular litter types raises two related questions. First, how do larvae find widely dispersed patches of food on the stream bottom? Second, how do the larvae decide whether, and how long, to remain on a given food patch before moving on to seek another? Neither visual nor olfactory cues appear to be used to find food. In the field, marked larvae rarely found patches of conditioned leaf litter when placed 30 cm downstream or upstream from them, although the leaves were readily colonized by unmarked, resident larvae [22]. In laboratory experiments, larvae appeared to move about randomly until they encountered a litter patch. Gustatory cues after feeding had begun then determined whether the larva left a patch or remained to continue feeding [25].
The second question is more complex. A larva that arrives at a potential food patch such as a decaying leaf and samples the material must then decide whether to remain on that patch or seek out another, potentially better patch. The simplest decision would be to remain on the patch if sampling suggests the material is palatable and nutrient content is sufficient to sustain growth, and abandon the patch if they are not. In choice experiments, a binary decision should lead to similar densities of larvae on any litter type which sustains growth (e.g., maple and alder) regardless of any qualitative differences between them. Poor quality litter such as oak should be quickly deserted.
Optimal foraging theory suggests larvae should remain on an acceptable patch of litter until the benefits of remaining no longer outweigh the cost of finding a new patch [26]. If patches vary widely in nutritive value, however, staying on the first patch which provides a minimum level of nutrition may not be the optimal strategy. Natural selection may favour larvae that abandon patches before they are exhausted to seek out a new and potentially better patch. In a temporally changing landscape in which litter patches appear and disappear and change quality over time, a “restless” larva may be better adapted than one that stays on the first acceptable patch. Further, if larvae do abandon acceptable patches, fidelity to a given patch should be greater when its nutritional value is greater, because the chance of finding a new, better patch is lower. In choice experiments, this behaviour would lead to larvae being distributed across samples of varying nutritional quality, with the greatest number remaining on the nutritionally best litter.
The project reported here provides information relevant to both these questions. Field experiments presented P. guttifer larvae with leaf litter from a range of riparian tree species to see if they distinguished among them. Complementary laboratory experiments determined whether individual larvae were more or less likely to abandon litter of different types.

2. Materials and Methods

2.1. Pycnopsyche Guttifer Walker

Pycnopsyche guttifer is the most common and wide-ranging species in the genus Pycnopsyche [27], which comprises 15 species of large (up to 3 cm long), detritus-feeding caddisflies. The larvae are common in slow water and the pools of cool, deciduous-woodland streams and rivers of eastern North America. Analyses of gut contents suggest that P. guttifer larvae consume leaf litter almost exclusively [28]. As with many leaf-shredding insects, the life cycle of Pycnopsyche appears to be synchronized with the annual pattern of litter fall. Eggs hatch in September or October when leaves fall from bankside trees, providing a food source for the early-instar larvae [29]. Early instars build portable cases of leaf discs or bark and twigs, but final-instar larvae (those observed here) make cases of wood, sometimes with rock fragments [27]. The case provides protection against predators and enhances water flow over the gills as the larvae move about in search of patches of detritus [30]. P. guttifer builds characteristic cases with protruding twigs oriented longitudinally [31], possibly as rudders. In summer, final-instar Pycnopsyche larvae enter aestivation for several weeks before pupation [32]. Adults emerge between July and October, depending on species and location [27].

2.2. Study Sites

Northern mainland Nova Scotia is drained by a network of small streams feeding into a set of short rivers which run in parallel drainages into the sea. The rivers run through a rural landscape dominated by short-rotation, boreal transition forest (northern hardwoods, pine, spruce) interrupted by hayfields, clear-cuts and country residences. There are few large settlements. River systems are gently sloped (1–4%) through most of their lengths and usually bear a closed-canopy riparian zone of largely deciduous trees. The region has a modified continental climate, with long, snowy winters and short, cool summers; annual precipitation is about 1300 mm [33].
Field experiments were done in the South River, a typical broad, shallow, regional river. The South River at Cummings Mills Cross Road (45°28′14″ N, 61°56′23″ W) averages 15.0 ± 1.4 m wide (n = 3) and is only partially shaded by a riparian forest of white ash (Fraxinus americana L.) and red maple (Acer rubrum L.). Most of the site is a run, with water 15–30 cm deep flowing over a substratum of cobble and rock. Typical of streams of this region, water in the South River is very soft, well oxygenated, and circumneutral to mildly acidic (Table 1). Nutrient concentrations are always low [34].
Polson’s Brook (45°27′00” N, 61°54′37” W), a tributary of the South River, is described in [22]. Polson’s Brook flows through upland mixed forest with some speckled alder (Alnus incana (L.) Moench) along the banks. The channel is 5–6 m wide at the study site; substratum is a mix of cobble, gravel and sand, with a few larger rocks. Polson’s Brook was used as an additional source of Pycnopsyche for laboratory experiments.
Discharge is relatively stable in Polson’s Brook, but rises quickly in the South River following rainstorms. Water temperatures (Table 1) are similar at both sites in spring. Pycnopsyche guttifer is common at both sites in spring.

2.3. Field Experiments

The field experiments tested the hypothesis that the number of Pycnopsyche larvae colonizing samples of conditioned litter after a fixed amount of time would differ among litter types, with the larvae showing consistent preferences for some litter types over others. Freshly fallen leaf litter of speckled alder, red maple, trembling aspen (Populous tremuloides Michx.) and red oak (Quercus rubra L.) were collected at leaf fall (late October) in previous years and stored air-dried. We chose these four species because they are common components of regional forest that differ in chemical composition, decomposition rate, and attractiveness to shredders. Leaf litter from N-fixing alder is very high in nitrogen [35] and is generally preferred by shredders, as noted earlier [16,17,18,19] and in our own work [22,36]. Oak leaf litter, at the other extreme, is noted for its tough structure and very slow decomposition rates [36,37,38] and did not attract Pycnopsyche in earlier work [22]. Litter of red maple and trembling aspen lie between these two extremes; the relatively low N content and slow decomposition rate of aspen litter [39] were expected to make it less attractive to Pycnopsyche than maple litter.
In Experiment 1, we weighed 12 samples of the four litter types, each weighing 2.00–2.03 g, to the nearest 0.001 g, remoistened them in a humid environment for 48 h to reduce brittleness, and then enclosed them in flat, numbered, 13 × 25 cm, mesh bags with a mesh size of 9 mm. Beginning 1 May 2021, the bags were suspended above the substratum in a small stream lacking shredders and allowed to condition for four weeks.
On 31 May, four bags of each type were laid across the stream bed in a 4 × 4 matrix in which one bag of each type occurred in each row. The arrangement of each row was drawn at random from the pool of 24 possible arrangements, without repeating. Bags were placed 50 cm apart in rows 2 m apart, with the intention that larvae could easily discover litter bags in the same row but were less likely to find other rows. Each bag was secured with a rock on the front edge. After 24 h, we carefully lifted each bag while sliding a small net underneath to catch any larvae that dropped off. Most larvae were on the underside of the bags. We counted the number of Pycnopsyche larvae on each bag, then returned them to the stream bed. Previous research showed that colonization is finished in 24 h [22]. We checked the bags for other conspicuous caddisfly shredders such as Oecetis or Lepidostoma, but found none. Then we returned the bags to the river and repeated the experiment in a different location within the wide riverbed. We ran the experiment six times in total (for a total of 24 replicates of each litter type), replacing the litter bags after the third run.
Preliminary data analysis revealed an unexpected effect of row position on the count of Pycnopsyche larvae. Consequently, data (log-transformed for normality) were analyzed using a factorial ANOVA, blocking on Row, with Day as the replication variable. Tukey’s test was used to distinguish among individual litter types.
Experiment 2 was designed to differentiate feeding preferences between pairs of litter types. Litter bags prepared as in the first experiment were conditioned for three weeks in May 2022. On 24 May, six bags of two litter types each were placed on the substratum of South River, 50 cm apart, in an alternating pattern, creating a row of 12 bags. A second row was laid 1 m downstream from the first row, with the bags alternating in the opposite sequence. After 24 h, the number of Pycnopsyche larvae colonizing the bags were counted as in the first experiment. The bags were then replaced on the substratum in a different order, with the same overall pattern, in a new location a few metres away. This experiment was repeated twice for Aspen vs. Maple (24 replicates total), twice for Aspen vs. Oak and three times for Alder vs. Maple (36 replicates). Statistical analysis of log-transformed data followed the same approach as in the first experiment. All statistical analysis used Statistix® v. 10.

2.4. Laboratory Experiments

Laboratory experiments were designed to test the second hypothesis, that Pycnopsyche larvae actively move from one food patch to another, sometimes deserting patches of litter even when they still provide food. In this scenario, the probability of leaving any given litter patch in a given time frame is inversely proportional to the quality of the food, which in turn should correlate with preferences shown in the field. Therefore, the larvae should abandon oak leaves almost immediately and alder rarely, with aspen and maple between the extremes.
Laboratory experiments used a 100 L recirculating indoor stream, consisting of a 60 L arena (63 cm × 126 cm) underlain by dark aquarium gravel, and a 40 L reservoir. A pump recirculated water at 4.3 L/min from the reservoir to the far end of the arena, where a diffuser distributed the inflow, maintaining a uniform depth of 8 cm and a water velocity of 0.9 cm/s. Water drained back into the reservoir over a wide lip. The system was filled with water from nearby Brierly Brook, which is chemically similar to that from upper South River. Brook water was added as needed to compensate for evaporation, and 20 L was replaced every 3 d.
Water temperature in the indoor stream was maintained slightly below ambient (2023: 16.3 ± 2.2 °C, 2024: 17.3 ± 1.3 °C) with a chiller (Polyscience) which cooled the water in the reservoir to 15 °C. Except when measurements were being taken, the larvae were kept under dim light [40] provided by a 15 W incandescent lamp located 60 cm above the arena. The lamp produced a light intensity of 11 lux at the water surface, equivalent to light intensity ~20 min after sunset in June.
Most Pycnopsyche larvae used in these experiments were captured from the South River at Cummins Mills Crossing, the same location used for the field experiments. Late-season larvae came from Polson’s Brook. Larvae were transported to the laboratory (30 min) in buckets with some river water and sticks of wood for habitat. They were allowed to acclimate to the arena for 24 h under dim light, without food.
Laboratory experiments used the same four litter types as in the field experiments: speckled alder, red maple, trembling aspen and red oak. For each litter type, twelve coarse-mesh bags, each containing 3 g of air-dried litter were suspended in a small stream to condition for 3 wk as in the field experiments. Conditioned litter bags were stored in stream water at 4 °C until used.
Experiment 3, in May–June 2023, followed the movement of larvae from a Source litter type, on which the larvae were initially placed, to a Sink litter type place 50 cm upstream. First, all the 40–60 larvae in the arena were moved to a tray filled with stream water. Three litter bags of the Source litter type were placed in line across the arena, 12 cm from the downstream end, such that they spanned the entire width of the arena. Bags were held in place with three 8.5 cm terra cotta dishes (drainage trays for houseplants) placed at each end. Litter bags containing the Sink litter type were placed in the same orientation 50 cm farther toward the inflow end of the arena, but out of the range of surface turbulence caused by the diffuser. Larvae on the downstream Source litter could potentially sense olfactory cues from the Sink litter.
Thirty actively moving larvae were selected from the tray and placed on the Source litter bags, ten on each bag. After 2, 4, and 24 h, I counted the number of larvae attached to each litter bag, including those attached on the underside, with minimal disturbance. Larvae on the substratum under the bags were not included in the count. Numbers for all three litter bags of each type were combined. Observations captured the number of larvae remaining on the Source litter in the first few hours, and the distribution of larvae among the two litter types and the matrix after 24 h. The experiment was repeated with different combinations of Source and Sink litters, in random order, until all 12 possible combinations were exhausted.
Larvae not used for a given run were maintained in the reservoir with conditioned litter as food. At the end of each run, all larvae were returned to the arena for another 24 h of gut clearing before the next run began. Actively moving larvae for the next run were again chosen from the pool of 40–60 animals. After two or three runs all larvae were replaced by a new set captured from the South River or Polson’s Brook. Litter bags were replaced after three runs.
Laboratory experiments were possible only during a brief window in May–June, after the spring freshet but before the larvae entered summer aestivation [32]. In 2024, we repeated tests with four litter combinations which were completed close to the end of the season in 2023 or had provided ambiguous results. This continuation used the same methods as in the previous year. Analysis used combined data from both years.
To test whether larvae showed a preference for different litter types or tended to move more in one direction (upstream or downstream) I used the final densities on all litter types after 24 h in all runs as replicates in a factorial ANOVA on litter type and location (Source downstream or Sink upstream). If larvae choose to leave a litter of low food quality, then the number in the matrix after 24 h (time for completion of colonization in the field) should be correlated with the quality of the food, as indicated by preferences in the field. To test this possibility, I assigned each of the litter types a rank indicator of quality from 1 (alder, most preferred) to 4 (oak, least preferred). I then computed indices of food preference in any given run by adding or multiplying the ranks of the two litter types provided. I computed linear regressions of number of larvae remaining in the matrix against food preference indices.
Experiment 4 (2025) followed movement of individual, marked larvae to see (1) whether they moved directionally while searching and (2) whether they stayed on or deserted leaf litter once they had discovered it. The arena was subdivided into a 5 × 10 matrix of 12.6 cm squares, marked along the perimeter of the basin. Litter bags containing conditioned litter was placed across the channel as in the previous experiment, occupying columns 2 (downstream) and 9 (upstream). Five or six large, active larvae were marked with coloured thread tied around the case or with a single drop of fast-drying nail polish. After marking, larvae were returned to the holding tray with the others until they began to move again. Then, the marked larvae were placed in the centre of the arena, a few centimetres apart, equidistant from the two litter types.
At intervals of 0.5, 1, 1.5, 2, 3, 4, 8 and 24 h, the cell in the matrix occupied by each larva was recorded. Observations were concentrated in the first few hours to detect random or directional initial movement and because larvae were expected to move less once they began to feed. (The larvae were indiscernible in video images.) The experiment was repeated six times, in random order, using a pair of dissimilar litter types (alder, oak), a pair of similar, favoured litter types (alder, maple) and a pair of disfavoured litter types (oak, aspen). Each combination was repeated with the upstream–downstream positions of the litter bags reversed. Marked larvae were used only once. A new pool of larvae were drawn from the South River or Polson’s Brook after every second run.
To test whether larvae move directionally in the arena, I defined cells in columns 1–4 as “downstream” and 7–10 as “upstream”. I then counted the number of times in each of the six runs that a larva was observed in these columns at 0.5 h and 1 h from the beginning. Larvae could occupy any cell in the arena within an hour. Larvae that remained in columns 5–6 were not included. I used a chi-square test to see whether the distribution of larvae in quadrants of the arena departed from random. I then repeated the calculation for lateral movement, comparing rows A–B against rows D–E.
Movement patterns revealed by the cell locations of the larvae were examined to see whether the larva stayed on or abandoned litter of each kind once they had discovered it. I noted whether larvae moved from one litter type to another, how many were on each litter type, or in the matrix, after 24 h, and how many observations found larvae in the matrix. Because larvae could not always be found, the number of total observations varied from one run to another.

3. Results

3.1. Field Experiments

Densities of Pycnopsyche on litter bags were lower in 2021 than in 2022, perhaps because the experiment was carried out in June, close to the summer aestivation period. The four-type choice experiment (Experiment 1) produced an unexpectedly strong effect of Row in the 4 × 4 matrix on colonization by Pycnopsyche (F3,81 = 11.3, p < 0.001). The most upstream row of bags always attracted the most larvae, with diminishing numbers in each row downstream (Figure 1). On average, litter bags in the first row attracted four times more larvae than those in the fourth row. Evidently larvae colonize disproportionately from upstream. ANOVA comparing litter type preferences therefore blocked against Row.
The four-type choice experiment demonstrated a significant effect of litter type (F3,69 = 6.5, p =0.006), but the separation of litter types was ambiguous. Larvae showed a clear preference for alder litter, while aspen and oak were identically disfavoured (Figure 2) and red maple was intermediate. There are few clear distinctions however, and the data are marked by high variation among individual bags. Densities on alder bags, for example, varied from 0 to 18, while those on oak ranged 0–7.
The two-type choice (Experiment 2) in 2022, when Pycnopsyche were more abundant, provided clearer results. In a choice between leaf litter of red maple and trembling aspen, counts of larvae on maple were an order of magnitude greater than on aspen (Figure 3A), a highly significant difference (F1,19 = 24.1, p = 0.0001). In turn, significantly more larvae colonized aspen litter than oak litter (F1,29 = 7.0, p = 0.013). The oak litter attracted very few larvae (Figure 3B). Finally, when two nutritious litter types, red maple and speckled alder, were compared, significantly more larvae colonized alder litter than maple (F1,29 = 14.3, p = 0.0007). About three times more larvae were counted on alder litter bags than on maple litter bags (Figure 3C). Taken together, the field data confirm that the order of preference of Pycnopsyche larvae feeding on leaf litter was alder > maple > aspen > oak. However, the attractiveness of maple litter depended on the litter type with which it was compared. The number of larvae on maple bags when paired with aspen was roughly four times greater than the count when paired with alder (Figure 3), and indeed exceeded the count on alder when it was paired with maple.

3.2. Laboratory Experiments

In Experiment 3, with few exceptions, half or more of the thirty larvae placed on a Source litter type moved into the matrix within the first four hours (Figure 4). The only exceptions among the 16 runs were two runs with maple as the Source and one with alder. In those runs, the number of larvae on the Source litter still diminished after 24 h. Larvae in the matrix appeared to move randomly upstream or downstream, sometimes encountering either the Sink litter or the Source again. Hence, final numbers on the Source or Sink litter samples and in the matrix after 24 h were the most useful for comparison of feeding preferences. Comparisons with the same Source and Sink litter types repeated in 2023 and 2024 produced similar patterns of larval movement.
In most runs, a net movement of larvae away from a less-favoured litter toward a more-favoured litter is apparent after the initial movement of larvae into the matrix (Figure 4). However, in no instance did all the larvae migrate to one litter type in 24 h, and in no instance did all the larvae leave the matrix in that time. The data suggest that different litter types were more or less favoured by the larvae, as expected, and that final densities on a particular litter type were independent of whether the litter was a Source or a Sink (Figure 4). ANOVA showed a highly significant effect of litter type (F3,24 = 11.5, p = 0.0001), but no effect of location (F1,24 = 0.01, p > 0.2), nor any significant interaction (F3,24 = 0.4, p > 0.2). Tukey’s test separated alder and maple litters from aspen and oak, but could not distinguish between members of each pair. The order of the means repeats that which was observed in the field (Figure 5).
Of the original 30 larvae placed on the Source litter type, the number remaining in the matrix after 24 h varied from 1 (maple and alder) to 21 (aspen and oak). Both the additive and multiplicative indices of relative food preference are significantly correlated with the number of larvae in the matrix after 24 h (additive: F = 16.7, p = 0.0011, R2 = 0.54; multiplicative: F = 22.0, p = 0.0004, R2 = 0.61, N = 16), although the latter is stronger (Figure 6).
In Experiment 4, marked larvae moved actively around the arena, again with no apparent selection for upstream or downstream. Movement was not a random walk. Rather, the larvae crawled in a more or less straight line until they encountered leaf litter or the wall of the arena, at which point they either stayed on the litter to feed or crawled away in a new direction. Observations 30 min apart sometimes encountered the larvae at opposite ends of the arena (Figure 7). Larvae sometimes could not be located at one or more observation times, probably because they were beneath litter, but all re-appeared by the end of the 24 h run. One larva died. The distribution of larvae after 0.5 and 1 h did not differ from random according to the chi-square test (upstream–downstream: χ2 = 5.9, df = 5, p > 0.20; lateral: χ2 = 6.6, df = 5, p > 0.20).
Actual paths followed by the larvae may be more complex than illustrated in Figure 7 because later observations were infrequent. When presented with alder and oak litter, a high-preference and low-preference pair, larvae crossed or abandoned oak litter frequently, but not necessarily immediately. Larvae on oak litter were observed there for one to five consecutive observations (mean 2.2 ± 1.5, n = 10) and always deserted the litter before 24 h. Larvae that encountered alder litter strongly tended to remain there: one larva crossed the alder litter twice; and another abandoned the alder litter after 1.5 h and returned after 4 h. All the other larvae remained on alder litter from when they encountered it. At the end of the 24 h observation, all the larvae were on alder litter. A total of 21 observations of 72 (29%) found larvae in the matrix.
When larvae were presented with alder or maple litter, both preferred food sources, larvae frequently deserted one litter type in favour of the other. Three larvae out of 10 remained on maple litter when they discovered it; one larva remained on alder litter. Two larvae abandoned alder litter before 0.5 h and 8 h, before returning again within 24 h. The remaining four larvae moved from maple to alder litter (two larvae), from alder to maple litter (1 larva) or from alder to maple and back to alder again (one larva). At the end of the 24 h run, four larvae out of 10 were on maple litter and six were on alder litter. A total of 21 observations of 60 (35%) found larvae in the matrix.
When presented with aspen and oak litter, both unpreferred food sources, larvae crossed and abandoned both litter types frequently. On average, larvae observed on aspen litter remained there for 1.9 ± 1.6 consecutive observations (n = 12), while those on oak remained for 1.4 ± 0.74 observations (n = 8). No larvae remained on either litter type from when they encountered it until the end of the run. After 24 h, three larvae of 10 were on oak litter, two were on aspen litter, and the remaining five were in the matrix. A total of 30 observations of 64 (47%) found larvae in the matrix.

4. Discussion

Densities of P. guttifer on litter bags in the first field experiment here, with four litter types, were comparable with those observed in earlier work. A survey of regional rivers [22] reported a median density of 3.5 animals/bag for maple litter and 8.5 animals/bag for alder litter, while a time-series experiment with alder litter produced ~ five animals/bag after 24 h. Densities in the two-species choice experiment the following year were substantially greater, however (Figure 3). The latter densities are well above the threshold where dramatic acceleration of decomposition rates appear.
As expected, the field experiments demonstrated that P. guttifer has definite preferences among food sources, after an equal duration of conditioning, in the order alder > maple > aspen > oak. This order matches the presumptive order of food quality based on nutritional content and overall decomposition rates for these four litter types [34,35,36,37,38,39]. Our results also correspond with previous work in which a strong preference for Alnus spp. has been almost universally observed [16,17,18,19,36,41].
Significantly, however, the variation within litter types was very high, often sufficient to obscure differences among them. Some bags of alder litter, in particular, always attracted substantially more larvae than others in the same set. This preference remained regardless of where the “popular” bags were placed in the 4 × 4 matrix. Presumably these bags contained litter on which fungal populations were more abundant or softening of tissue was more advanced. Replacing the litter bags with fresh ones at intervals during the experiment reduced the bias induced by popular bags, but their presence illustrates that feeding preferences are dictated by characteristics of a particular patch of food, not only the general litter type.
The field experiments also demonstrated a strong and unexpected directional movement of the larvae. Far more larvae colonized the most upstream bags in the matrix, suggesting that larvae found the bags predominantly by moving downstream. This finding contradicts both our laboratory results and the findings of Motyka et al. [25], all of which show non-directional movement of larvae. The discrepancy probably arises from the high water velocity in the South River (Table 1) where the field experiments were performed, compared with the slow movement (0.01 m/s) in the laboratory. For a relatively large, unflattened larvae such as Pycnopsyche, moving downstream by crawling or rolling with the current would probably be energetically favourable to moving against the current upstream. Lancaster et al. [42] observed that larvae of the limnephilid caddisfly Potamophylax latipennis also moved predominantly downstream, especially in strong currents. However, Cummins [32] observed that P. guttifer moved predominantly upstream in laboratory flumes with a current velocity of only 0.03 m/s, close to the mean water velocity Cummins observed in the field. This observation, in turn, highlights that the congregations of Pycnopsyche observed in the South River here and in earlier work [22] are anomalous in that the species is typically found in slow backwaters and pools where organic matter collects [27,32]. In its more typical habitat, random searching would be expected. Furthermore, the preference to colonize litter from upstream again demonstrates that the larvae do not use olfaction to find food patches.
The attraction of maple leaf litter to Pycnopsyche larvae appears to depend on the other food sources around it. Far more larvae colonized maple when the nearby choice was aspen litter, rather than alder litter. Yet, the total counts of animals on both litter types in the two comparisons were not very different. This pattern suggests that, while some larvae encountered maple litter directly, others may have discovered the aspen litter but abandoned it to continue searching, eventually finding the nearby maple litter. In contrast, when alder and maple were paired, some animals appear to have left the maple litter to colonize the nearby alder. This migration occurred despite the fact that maple is a preferred litter type in other contexts, and attracted most larvae when aspen was nearby.
These observations support two hypotheses. The first is that the larvae can sense the presence of alder litter, or of a highly preferred food source in general, and move toward it. It is unclear, however, how the information would be received. Olfaction would be the most likely candidate, but that would not work from upstream, in which direction most colonization occurred, or laterally. Larvae in the laboratory did not distinguish between food sources upstream or downstream. It is unlikely that the larvae can see litter at a distance in the complex environment of a stream bottom, or recognize what it is, from a distance, inside a mesh bag.
The second, more likely hypothesis, is that larvae periodically abandon food patches in search of better ones, even when the current food choice is acceptable. If larvae remained on the first patch of acceptable food, then densities on alder and maple should be equal, and the density on maple should be independent of the litter types around it.
Laboratory experiments support results from the field, with some variation related to the artificial habitat. In the field, larvae placed on bags containing conditioned litter in flowing water stayed there for at least the next hour [22]. In the laboratory, however, most larvae deserted the litter bags they had been placed on within an hour, regardless of the litter type, perhaps as an escape response after being handled. Some larvae then returned to the litter over the following day of observations. Nevertheless, over the duration of the experiment, larvae demonstrated the same order of preference among litter types as in the South River, although again variability between bags obscured the general trend. Moreover, the various litter types attracted the same numbers of larvae whether placed as a downstream, Source litter or an upstream Sink.
Significantly, the number of larvae on any litter type after 24 h were always less than the total in the arena. Some larvae were always in the matrix, fewer when alder was available, and more as the subjective food quality declined. In tests with aspen and oak, larvae spent almost half the time in the matrix. It seems unlikely that the larvae, which tended to move actively, would have failed to discover food sources in the comparatively small arena within 24 h. It appears, therefore, that some larvae deserted food sources once they had found them. That the number of larvae in the matrix after 24 h could be predicted with coarse accuracy from a simple, ad hoc index of food preference strongly suggests that the tendency of larvae to desert a food patch depends inversely on the perceived quality of the food source.
Finally, Experiment 4 confirms, first, that larvae move about actively, but choose directions randomly. Larvae frequently traversed the length of the arena more than once in 24 h, once three times, and may have done so more often in the long periods between later observations. Yet, an analysis of early positions could not detect any sign that larvae preferentially crawled downstream or upstream, or in either lateral direction. It appears then, that as Motyka et al. [25] observed, larvae randomly search for food patches and upon arrival use gustatory cues to judge the quality of the litter.
This experiment also confirms that larvae frequently abandon litter patches once they have found them. The fidelity to any given litter type depends on its apparent nutritional quality, but even alder is sometimes abandoned. Hence, when paired with an oak litter, a low-preference source, alder tends to accumulate all the larvae over time. When paired with maple, however, more equal attraction combined with random searching leads to larvae sorting between the two litter types. In the third situation, with two, low-preference litter types, more larvae were observed in the matrix on most occasions, as the larvae readily abandoned poor food in search of something better.
The major limitation of this research is that the P. guttifer larvae may not behave identically in a river as in a highly artificial laboratory arena. Indeed, the observation that larvae tend to move off litter bags in the laboratory but stay on them in the field (perhaps because of water current) illustrates that the field situation is not perfectly emulated in the laboratory. Also, larvae in the arena may have used litter bags as shelter and habitat in the otherwise featureless matrix, although the frequency with which they deserted aspen and oak suggests that food was the main attraction. Nevertheless, general patterns of behaviour were congruent between field and laboratory, especially the order of preference of the four litter types tested. Because the work was done with late-instar larvae, some animals may have stopped feeding while experiments were underway. Repetition of some runs in Experiment 3 were intended to address this possibility. Unfortunately, the highly seasonal nature of discharge in regional rivers means that larvae cannot be procured nor can field experiments proceed until the rivers are low enough to be safe. Feeding habits of P. guttifer over the entire larval period have not been explicated.
Summarily, these experiments support the “restless larvae” hypothesis which suggests that larvae of Pycnopsyche guttifer, and perhaps other litter-feeding stream insects, periodically move away from litter patches in search of new ones, even when the food patch they are on is not exhausted. Preferred litter types, presumably more nutritious or palatable, appear to retain the larvae longer than less attractive alternatives. A larva that deserts a food patch occasionally (high-quality litter) or frequently (low-quality litter) in search of a better food source may optimize its energy intake. Whether these findings apply to Pycnopsyche in other regions, to other leaf-shredding caddisflies, or indeed to any litter-feeding stream invertebrate, are important questions for future research.

Funding

This research received no external funding.

Data Availability Statement

All data reported here are available from the author.

Acknowledgments

Thanks to Melissa Pacheco Checa, Evelyn Pequenera-Griffin and Thomas MacLellan for help in the laboratory and field, and to colleague Tammy Rodela for help with monitoring larval insects in the laboratory. Russell Wyeth provided advice on insect behaviour. Comments from three reviewers and a journal editor substantially improved the original manuscript. St. Francis Xavier University provided laboratory space for this research.

Conflicts of Interest

The author declares no conflicts of interest.

References

  1. Webster, J.R.; Benfield, E.F. Vascular plant breakdown in freshwater ecosystems. Annu. Rev. Ecol. Syst. 1986, 17, 567–594. Available online: https://www.jstor.org/stable/2097009 (accessed on 15 April 2026). [CrossRef]
  2. Wallace, J.B.; Eggert, S.L.; Meyer, J.L.; Webster, J.R. Stream invertebrate productivity linked to forest subsidies: 37 stream-years of reference and experimental data. Ecology 2015, 96, 1213–1228. [Google Scholar] [CrossRef] [PubMed]
  3. Riis, T.; Kelly-Quinn, M.; Aguiar, F.C.; Manolaki, P.; Bruno, D.; Bejarano, M.D.; Clerici, N.; Fernandes, M.R.; Franco, J.C.; Pettit, N.; et al. Global overview of ecosystem services provided by riparian vegetation. BioScience 2020, 70, 501–514. [Google Scholar] [CrossRef]
  4. Graça, M.A.S. The role of invertebrates on leaf litter decomposition in streams—A review. Int. Rev. Hydrobiol. 2001, 86, 383–393. [Google Scholar] [CrossRef]
  5. Hieber, M.; Gessner, M.O. Contribution of stream detritivores, fungi, and bacteria to leaf breakdown based on biomass estimates. Ecology 2002, 83, 1026–1038. [Google Scholar] [CrossRef]
  6. Sontanja, M.; Pellan, L.; Piscart, C. Macroinvertebrate identity mediates the effects of litter quality and microbial conditioning on leaf litter recycling in temperate streams. Ecol. Evol. 2018, 8, 2542–2553. [Google Scholar] [CrossRef] [PubMed]
  7. Marks, J.C. Revisiting the fate of dead leaves that fall into streams. Annu. Rev. Ecol. Evol. Syst. 2019, 50, 547–558. [Google Scholar] [CrossRef]
  8. Jonsson, M.; Sponseller, R.A. The role of macroinvertebrates on plant litter decomposition in streams. In The Ecology of Plant Litter Decomposition in Stream Ecosystems; Swan, C.M., Boyero, L., Canhoto, C., Eds.; Springer: Berlin/Heidelberg, Germany, 2021. [Google Scholar] [CrossRef]
  9. Chung, N.; Suberkropp, K. Effects of aquatic fungi on feeding preferences and bioenergetics of Pycnopsyche gentilis (Trichoptera: Limnephilidae). Hydrobiologia 2009, 630, 257–269. [Google Scholar] [CrossRef]
  10. Gessner, M.O.; Chauvet, E.; Dobson, M. A perspective on leaf litter breakdown in streams. Oikos 1999, 85, 377–384. [Google Scholar] [CrossRef]
  11. Bärlocher, F.; Kendrick, B. Leaf-conditioning by microorganisms. Oecologia 1975, 20, 359–362. [Google Scholar] [CrossRef] [PubMed]
  12. Golladay, S.W.; Webster, J.R.; Benfield, E.F. Factors affecting food utilization by a leaf shredding aquatic insect: Leaf species and conditioning time. Holarct. Ecol. 1983, 6, 157–162. [Google Scholar] [CrossRef]
  13. Bastian, M.; Boyero, L.; Jackes, B.R.; Pearson, R.G. Leaf litter diversity and shredder preferences in an Australian tropical rain-forest stream. J. Trop. Ecol. 2007, 23, 219–229. [Google Scholar] [CrossRef]
  14. Heller, R.E.; Stouffer, A.N.; Iyengar, E.V. Effects of species of leaves and conditioning time on vernal colonization by temperate lotic isopods (Lirceus sp.). Hydrobiology 2024, 3, 63–73. [Google Scholar] [CrossRef]
  15. Rumbos, C.I.; Stamopoulos, D.; Georgoulas, G.; Nikolopoulou, E. Factors affecting leaf litter decomposition by Micropterna sequax (Trichoptera: Limnephilidae). Int. Rev. Hydrobiol. 2010, 95, 383–394. [Google Scholar] [CrossRef]
  16. Benfield, E.F.; Webster, J.R. Shredder abundance and leaf breakdown in an Appalachian Mountain stream. Freshw. Biol. 1985, 15, 113–120. [Google Scholar] [CrossRef]
  17. Irons, J.G.; Oswood, M.W.; Bryant, J.P. Consumption of leaf detritus by a stream shredder: Influence of tree species and nutrient status. Hydrobiologia 1988, 160, 53–61. [Google Scholar] [CrossRef]
  18. Motomori, K.; Mitsuhashi, H.; Nakano, S. Influence of leaf litter quality on the colonization and consumption of stream invertebrate shredders. Ecol. Res. 2013, 16, 173–182. [Google Scholar] [CrossRef]
  19. Balibrea, A.; Ferreira, V.; Gonçalves, V.; Raposeiro, P.M. Consumption, growth and survival of the endemic stream shredder Limnephilus atlanticus (Trichoptera, Limnephilidae) fed with distinct leaf species. Limnologica 2017, 64, 31–37. [Google Scholar] [CrossRef]
  20. Arsuffi, T.L.; Suberkropp, K. Selective feeding by stream caddisfly (Trichoptera) detritivores on leaves with fungal-colonized patches. Oikos 1985, 45, 50–58. [Google Scholar] [CrossRef]
  21. Rong, Q.; Sridhar, K.R.; Bärlocher, F. Food selection in three leaf-shredding stream invertebrates. Hydrobiologia 1985, 316, 173–181. [Google Scholar] [CrossRef]
  22. Porter, J.D.; Taylor, B.R. Abrupt population increase by the leaf-shredding caddisfly Pycnopsyche guttifer profoundly changes leaf litter decomposition in streams of eastern Canada. Fund. Appl. Limnol. 2021, 95, 153–171. [Google Scholar] [CrossRef]
  23. Richardson, J.S. Seasonal food limitation of detritivores in a montane stream: An experimental test. Ecology 1991, 72, 873–887. [Google Scholar] [CrossRef]
  24. Graça, M.A.S.; Ferreira, R.C.F.; Coimbra, C.N. Litter processing along a stream gradient: The role of invertebrates and decomposers. J. N. Am. Benthol. Soc. 2001, 20, 408–420. [Google Scholar] [CrossRef]
  25. Motyka, G.L.; Merritt, R.W.; Klug, M.J.; Miller, J.R. Food-finding behavior of selected aquatic detritivores: Direct or indirect behavioral mechanism? Can. J. Zool. 1985, 63, 1388–1394. [Google Scholar] [CrossRef]
  26. King, A.J.; Marshall, H.H. Optimal foraging. Curr. Biol. 2022, 32, R680–R683. [Google Scholar] [CrossRef] [PubMed]
  27. Wiggins, G.B. Larvae of the North American Caddisfly Genera (Trichoptera), 2nd ed.; University of Toronto Press: Toronto, ON, Canada, 2000. [Google Scholar]
  28. Hutchens, J.J.; Benfield, E.F.; Webster, J.R. Diet and growth of a leaf-shredding caddisfly in southern Appalachian streams of contrasting disturbance history. Hydrobiologia 1997, 346, 193–201. [Google Scholar] [CrossRef]
  29. Mackay, R.J.; Kalff, J. Ecology of two related species of caddisfly larvae in the organic substrates of a woodland stream. Ecology 1973, 54, 499–511. [Google Scholar] [CrossRef]
  30. Morse, J.C.; Holzenthal, R.W.; Robertson, D.R.; Rasmussen, A.K.; Currie, D.C. Trichoptera. In An Introduction to the Aquatic Insects of North America, 5th ed.; Merritt, R.W., Cummins, K.W., Berg, M.B., Eds.; Kendall Hunt Publishing Company: Dubuque, IA, USA, 2019. [Google Scholar]
  31. Flint, O.S. Taxonomy and Biology of Nearctic Limnephilid Larvae (Trichoptera), with Special Reference to Species in Eastern United States; Cornell University: Ithaca, NY, USA, 1960; Volume 40, pp. 1–117. [Google Scholar]
  32. Cummins, K.W. Factors limiting the microdistribution of larvae of the caddisflies Pycnopsyche lepida (Hagen) and Pycnopsyche guttifer (Walker) in a Michigan stream. Ecol. Monogr. 1964, 34, 271–295. [Google Scholar] [CrossRef]
  33. David, D.S.; Browne, S. (Eds.) Natural History of Nova Scotia. Volume II: Theme Regions; Nimbus Publishing: Halifax, NS, Canada, 1996; 304p, Available online: https://ojs.library.dal.ca/index.php/NSM/issue/view/352 (accessed on 5 July 2026).
  34. Andrushchenko, I.V.; Taylor, B.R.; Toxopeus, J.; Wilson, E. Congregations of the leaf-shredding insect Lepidostoma togatum mediate exceptionally rapid mass loss from leaf litter in Nova Scotia rivers. Hydrobiologia 2017, 788, 245–265. [Google Scholar] [CrossRef]
  35. Huss-Danell, K. Nitrogen in shoot litter, root litter and root exudates from nitrogen-fixing Alnus incana. Plant Soil 1986, 91, 43–49. [Google Scholar] [CrossRef]
  36. Taylor, B.R.; Mallaley, C.; Cairns, J.F. Limited evidence that mixing leaf litter accelerates decomposition or increases diversity of decomposers in streams of eastern Canada. Hydrobiologia 2007, 592, 405–422. [Google Scholar] [CrossRef]
  37. Leroy, C.J.; Marks, J.C. Litter quality, stream characteristics and litter diversity influence decomposition rates and macroinvertebrates. Freshw. Biol. 2006, 51, 605–617. [Google Scholar] [CrossRef]
  38. Wymore, A.S.; Bourakovsky, A.; Fazeka, H.M.; Marks, J.C.; McDowell, W.H. Influence of watershed suburbanization on leaf litter decomposition and microbial activity. Hydrobiologia 2022, 849, 245–260. [Google Scholar] [CrossRef]
  39. Moore, T.R.; Trofymow, J.A.; Taylor, B.R.; Prescott, C.E.; Camire, C.; Duschene, L.; Fyles, J.; Kozak, L.; Kranabetter, M.; Morrison, I.; et al. Litter decomposition rates in Canadian forests. Glob. Change Biol. 1999, 5, 75–82. [Google Scholar] [CrossRef]
  40. Walbeek, T.J.; Harrison, E.M.; Gorman, M.R.; Glickman, G.L. Naturalistic intensities of light at night: A review of the potent effects of very dim light on circadian responses and considerations for translational research. Front. Neurol. 2021, 12, 625334. [Google Scholar] [CrossRef] [PubMed]
  41. Jacobsen, D.; Friberg, N. Food preference of the Trichopteran larva Anabolia nervosa from 2 streams with different food availability. Hydrobiologia 1995, 308, 139–144. [Google Scholar] [CrossRef]
  42. Lancaster, J.; Buffin-Bélanger, T.; Reid, I.; Rice, S. Flow- and substratum-mediated movement by a stream insect. Freshw. Biol. 2006, 51, 1053–1069. [Google Scholar] [CrossRef]
Hydrobiology 05 00022 g001
Figure 1. Effect of row position on number of Pycnopsyche larvae colonizing four kinds of leaf litter in the South River, spring 2021. The litter bags were laid in a 4 × 4 matrix with 1 m between rows. Data are means for each row across all litter types (n = 6). Error bars are standard deviations.
Figure 1. Effect of row position on number of Pycnopsyche larvae colonizing four kinds of leaf litter in the South River, spring 2021. The litter bags were laid in a 4 × 4 matrix with 1 m between rows. Data are means for each row across all litter types (n = 6). Error bars are standard deviations.
Hydrobiology 05 00022 g001
Hydrobiology 05 00022 g002
Figure 2. Mean number of Pycnopsyche larvae after 24 h on litter bags containing deciduous leaf litter of different kinds. Error bars are standard deviations (n = 24). Litter types with different letters above are significantly different (Tukey’s Test, p < 0.05).
Figure 2. Mean number of Pycnopsyche larvae after 24 h on litter bags containing deciduous leaf litter of different kinds. Error bars are standard deviations (n = 24). Litter types with different letters above are significantly different (Tukey’s Test, p < 0.05).
Hydrobiology 05 00022 g002
Hydrobiology 05 00022 g003
Figure 3. Median number of Pycnopsyche larvae after 24 h on litter bags containing deciduous leaf litter of two kinds. Error bars are 25th and 75th percentiles. All differences are significant (Tukey’s Test, p < 0.05). Maple with aspen, n = 12; all others, n = 18. (A) Maple with Aspen; (B) Aspen with Oak; (C) Maple with Alder.
Figure 3. Median number of Pycnopsyche larvae after 24 h on litter bags containing deciduous leaf litter of two kinds. Error bars are 25th and 75th percentiles. All differences are significant (Tukey’s Test, p < 0.05). Maple with aspen, n = 12; all others, n = 18. (A) Maple with Aspen; (B) Aspen with Oak; (C) Maple with Alder.
Hydrobiology 05 00022 g003
Hydrobiology 05 00022 g004
Figure 4. Mean number of Pycnopsyche larvae (of 30) on each litter type over 24 h when used as a Source (open circles) or a Sink (closed circles) for larvae. Error bars are standard deviations (n = 3–5). Lower line offset 0.5 h for clarity.
Figure 4. Mean number of Pycnopsyche larvae (of 30) on each litter type over 24 h when used as a Source (open circles) or a Sink (closed circles) for larvae. Error bars are standard deviations (n = 3–5). Lower line offset 0.5 h for clarity.
Hydrobiology 05 00022 g004
Hydrobiology 05 00022 g005
Figure 5. Mean number of Pycnopsyche larvae (of 30) on litter of different types after 24 h in two-choice experiments in a laboratory arena. Data combine upstream and downstream positions in both 2023 and 2024. Error bars are standard deviations (n = 7–9). Litter types with different letters above are significantly different (Tukey’s test, p < 0.05).
Figure 5. Mean number of Pycnopsyche larvae (of 30) on litter of different types after 24 h in two-choice experiments in a laboratory arena. Data combine upstream and downstream positions in both 2023 and 2024. Error bars are standard deviations (n = 7–9). Litter types with different letters above are significantly different (Tukey’s test, p < 0.05).
Hydrobiology 05 00022 g005
Hydrobiology 05 00022 g006
Figure 6. Relationship between number of larvae (of 30) in the matrix after 24 h in two-choice experiments and the product of the rank orders of litter types by preference, as shown in Figure 5. Dotted line is the line of best fit (p = 0.0004, R2 = 0.61).
Figure 6. Relationship between number of larvae (of 30) in the matrix after 24 h in two-choice experiments and the product of the rank orders of litter types by preference, as shown in Figure 5. Dotted line is the line of best fit (p = 0.0004, R2 = 0.61).
Hydrobiology 05 00022 g006
Hydrobiology 05 00022 g007
Figure 7. Positions of some marked larvae in a laboratory arena with a choice of litter types over 24 h. One example (of 5 or 6) is given for each combination and position of litter tested. Numbers indicate time in hours after larvae were placed in the middle of the arena (black spot). Broken lines indicate times when the larvae could not be found.
Figure 7. Positions of some marked larvae in a laboratory arena with a choice of litter types over 24 h. One example (of 5 or 6) is given for each combination and position of litter tested. Numbers indicate time in hours after larvae were placed in the middle of the arena (black spot). Broken lines indicate times when the larvae could not be found.
Hydrobiology 05 00022 g007
Table 1. Summary of physical features of the study site on the South River at Cummings Mills Cross Road. Data are spot measurements taken at mid-morning in June 2021. Water depth was measured at 8–12 points among litterbags on each of 14 dates. Water velocity is the mean of 9 daily means, each based on the mean of 3–5 measurements.
Table 1. Summary of physical features of the study site on the South River at Cummings Mills Cross Road. Data are spot measurements taken at mid-morning in June 2021. Water depth was measured at 8–12 points among litterbags on each of 14 dates. Water velocity is the mean of 9 daily means, each based on the mean of 3–5 measurements.
VariableMeanStandard DeviationN
Temperature (°C)15.02.410
Conductivity (μS/cm)50.416.68
Dissolved Oxygen (mg/L)10.72.46
pH (units)7.20.279
Water Depth (cm)21.98.314
Water Velocity (m/s)0.340.149
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Taylor, B.R. Effect of Litter Type on Colonization and Desertion of Food Patches by a Litter-Shredding, Limnephilid Caddisfly. Hydrobiology 2026, 5, 22. https://doi.org/10.3390/hydrobiology5030022

AMA Style

Taylor BR. Effect of Litter Type on Colonization and Desertion of Food Patches by a Litter-Shredding, Limnephilid Caddisfly. Hydrobiology. 2026; 5(3):22. https://doi.org/10.3390/hydrobiology5030022

Chicago/Turabian Style

Taylor, Barry R. 2026. "Effect of Litter Type on Colonization and Desertion of Food Patches by a Litter-Shredding, Limnephilid Caddisfly" Hydrobiology 5, no. 3: 22. https://doi.org/10.3390/hydrobiology5030022

APA Style

Taylor, B. R. (2026). Effect of Litter Type on Colonization and Desertion of Food Patches by a Litter-Shredding, Limnephilid Caddisfly. Hydrobiology, 5(3), 22. https://doi.org/10.3390/hydrobiology5030022

Article Metrics

Back to TopTop