Spider samples were taken from DGFC's four permanent botanical plots and from the nearby Hillco Estate oil palm plantation (5.415°N 118.016°E) just across the Kinabatangan River. Two of the botanical plots (2 and 3) are subject to regular inundation (1–3 times per year) while the two remaining plots (1 and 4) are flooded only during extraordinary weather events (approximately once in 6–7 years). Despite inquiries, we were unable to determine details of the oil palm plantation history and management. Trees ranged from 2–3 m in height and we witnessed workers applying an unknown chemical.

Within the DGFC permanent botanical plots, 4–6 points were randomly selected; eight points from a block of oil palm plantation also were selected randomly. At each point, 1 m² was dusted for spider webs 0–10 cm above the ground using a corn starch puffer (Carico 1977) and as many spiders as possible were collected using a pooter. In the oil palm plantation, additional plotless sampling was conducted. All adult spiders from the plot sampling were counted and sorted; the plotless sample was checked for the presence of Crassignatha.

A total of 79 adult ground-web-building spiders were collected during the plot survey. Overall ground web spider density was significantly higher in the riparian forest (5.2 per m²) compared to the other habitats investigated (1.3 and 1.75 per m² in riverine forest and oil palm plantation, respectively; ANOVA with Tukey's pairwise comparisons, p < 0.05). Crassignatha danaugirangensis sp. n. was the most abundant species found with 38 adults (48.1%, 3.8 per m²). The next most abundant species overall was a member of the Hahniidae, with 7 individuals (8.9%). Crassignatha danaugirangensis was found only in the riparian forest plots. Plotless sampling in the oil palm plantation failed to find any Crassignatha. See also Suppl. material 1 for raw morphospecies abundance data by plot.

Periodic inundation is a regular feature of the forest at Danau Gurang. Some of the low lying forests, including botanical plots 2 and 3, were flooded one to two weeks prior to this study. In addition to Crassignatha danaugirangensis, the community of ground-web builders at Danau Girang includes linyphiids, mysmenids, theridiosomatids, and hahniids. The difference in ground web spider abundance found in the recently flooded and unflooded botanical plots can be attributed to the exclusive presence of Crassignatha danaugirangensis in the unflooded plots (Fig. 5, Table 1); reanalysis of the data without Crassignatha erases the difference in ground-web spider density between the habitat types (ANOVA, p = 0.88). This suggests that Crassignatha is particularly sensitive to forest disturbance, whether this is due to natural causes like flooding or profound anthropogenic causes such as palm oil agriculture.

Figure 5.  

Density of Crassignatha danaugirangensis sp. n. as a proportion of all adult ground-web-building spiders sampled from 1 m² plots in riparian forest, riverine forest, and oil palm plantation. See Suppl. material 1 for raw morphospecies abundance data.

As a coda to the field course, we organized a Crassignatha blitz. Students, instructors, and DGFC staff were organized into teams of two and sent to various points around the trail network. Each team used a sock to contain a quantity of corn starch, which when lightly tapped over the leaf litter, suffices as a puffer. Teams were trained to identify Crassignatha webs, and were asked to search for webs in their assigned area for 10 minutes. Teams were asked to photodocument a sample of the webs they observed, especially any observations about which they were not entirely certain. The results show that Crassignatha danaugirangensis is widespread along the DGFC trail network, but is rare if not absent in the most recently flooded forest patches. This raises the question: does Crassignatha danaugirangensis actually prefer riparian forest over riverine forest, or is this the case only shortly after an inundation event? To answer this question, further study of this species will be required. But in the absence of a durable and accessible taxonomy, it becomes virtually impossible to accumulate knowledge about a species between studies conducted by independent researchers. Until now, Crassignatha danaugirangensis was one of countless undescribed species. The ecological sensitivity of this species suggested by our brief study calls for further observation and monitoring. It also raises new questions about the inundation forest ecology of the invertebrates at a research station normally focused on some of the world's most charismatic vertebrates. The taxonomic description presented here and made accessible to all via an open access cybertaxonomic journal will facilitate this activity.