Fauna collections were carried out in two sites of Fossli Slot #2 and one site from Kiku Pot (see below the description of caves). Sampling was focused on regions where visible signs of chewed-up organic matter could be seen. Collections were made manually using #00 brushes and kept in small vials with 90% ethanol. Once the specimens were spotted, they were dabbed with a brush moistened with ethanol and immediately placed into vials, labelled and sealed for transport.

The specimens were washed using distilled water and were put between slides and glass coverslips to be examined under a phase-contrast optical microscope (Leica DMLS) using Marc André II solution. The illustrations were made with a drawing tube and the measurements were taken with an ocular micrometre. For measuring the body length, the specimens were mounted ‘in toto’ and were measured from the base of the frontal process distal macrochaetae to the abdomen’s supra-anal valve. Two paratypes were coated with palladium-gold used for scanning electronic microscopic photography (Hitachi S-4100) and measurement of the sensilla.

The morphological descriptions and abbreviations used in this article follow Condé (1956). We use gouge sensilla for the concavo-convexly shaped sensilla located on the antennae and described by Bareth and Condé (1981), the function of which is still unknown, and rosette gland formations for the epicuticular glands described in several Campodeinae species (Bareth and Juberthie-Jupeau 1996).

Silvestri (1911) described the genus, and Paclt (1957) redescribed it. By following these taxonomical criteria and adding a few others, it is worth attempting a new description of Haplocampa:

Antennae with moniliform antennomeres and short cerci with a few primary articles. Subtrapezoidal head slightly enlarged in the posterior lateral side. Medial anterior (1+1), lateral anterior (1+1) and lateral posterior (1+1) macrosetae on pronotum; medial anterior (1+1), lateral anterior (1+1) and lateral posterior (2+2) on mesonotum; and medial anterior (1+1) and lateral posterior (1+1-2+2) on metanotum. Dorso-femoral macrosetae one. One to three tibial-ventral macrosetae. Tarsus short and enlarged. Unequal claws with large lateral crests and no lateral process; basal portion of both claws with a visible spiny protrusion; posterior claw with a large backward overhang. Medial anterior or medial posterior urotergal macrosetae present; with or without lateral anterior urotergal macrosetae; and, lateral posterior urotergal macrosetae 0, 1 (lp₂) or 2 (lp_2,3) on urotergites V to VII. Urosternite I with 6+6 macrosetae; urosternites II to VII with 4+4 macrosetae; urosternite VIII with 1+1. Plain stylus with smooth or a few tiny barbs on stylus setae. Male with glandular a₁ and g₁-setae. Female with glandular a₁-setae.

Fossli #1 (L63.5 D25): swallet entrance with active surface stream. First sightings of H. wagnelli sp. n. on 5^th April 2005.

Fossli #2 (L67.5 D26.5): small slot entrance filled with rotten wood. Upper samples collected 15^th July 2018, 30 m from the entrance in a moist bed of sandy organic debris. Lower samples collected 6^th August 2018, 60 m from the entrance in a moist bed of sandy organic debris 2 m above sump. H. wagnelli sp. n. mud workings visible at both sites.

Kiku Pot cave, (L1489 D92): Breakdown entrance with active stream passage throughout system. H. wagnelli sp. n. first sightings August 2017 while surveying. Collection taken 5^th August 2018, at bottom waterfall just below where water disappears approximately 150 m from the entrance on a medium. A medium sized bed of sandy organic material with scattered woody debris. H. wagnelli sp. n. workings were visible over 75%. Area floods during high water. Many mud workings visible on muddy shelves higher in passage. At the time of collection, an adult fungus gnat was noticed hopping around.

Vancouver Island is known to have over 1600 caves, with more mapped and explored caves than the rest of Canada combined. Some caves reach over 10 km, and some host unique geological, palaeontological, archaeological and biological features. Vancouver Island caves are mostly active caves, with streams and rivers running through them most of the year. The caves help the streams to maintain constant water temperatures year round and a proper pH, which increases water quality for fish and wildlife. So far, little has been done to protect the caves from poor logging, mining and recreational practices. All caves west of the Port Alberni area are easily accessed by the public and none so far have protection or proper management. Unfortunately, some already have seen misuse. More needs to be done in the future if we want to protect this special resource.

In Haplocampa species, their unequal claws have large lateral crests, as do many genera of Plusiocampinae. Applying the right criteria, Paclt (1957) included Haplocampa within Campodeinae. In fact, many taxonomical features of Haplocampa are held by Campodeinae, such as the distribution and number of pronotum macrosetae, with at most 3+3 medial anterior, lateral anterior and lateral posterior macrosetae. Other notable features of Haplocampa are shared with many Campodeinae genera, namely, epicuticle with rosette gland formations; plain frontal process; number and distribution of urotergal macrosetae with medial macrosetae and a maximum of three pairs of lateral posterior macrosetae; simple stylus setae; and number and distribution of urosternal macrosetae, with up to eight pairs of macrosetae on first the urosternite, up to five pairs of macrosetae on the second to seventh urosternites and one pair on the eighth.

In all likelihood, Pacificampa Chevrizov, 1978 is the most closely related to Haplocampa, given the strong similarities in the number and distribution of the macrosetae body and the absence of lateral pretarsus processes, with one important difference: the lack of lateral crests in Pacificampa (Chevrizov 1978; Sendra et al. 2018). The same differential feature is shown by other related genera of Haplocampa, in Metriocampa Silvestri, 1912 and Eumesocampa Silvestri, 1933. All these four genera can be found in the East of Asia and in North America, on both sides of the northern Pacific Ocean, which suggests dispersal events over the Bering Land Bridge. This hypothetical palaeobiogeographical distribution was suggested by Ferguson (1997) for Eumesocampa, Haplocampa, Pacificampa and Plutocampa Chevrizov, 1978. Another perhaps more parsimonious hypothesis could be that they had an ancient Laurasian distribution and their current distribution reflects a vicariance event after the breacking of the Bering Bridge (Loris Galli pers. suggestion).

Haplocampa and Pacificampa share ecological similarities, and phylogenetic, too. Both are present in subterranean ecosystems. In the case of Pacificampa, all of the five known species have been found in caves and can be considered troglobites. Three of these species can be found in the north of the Korean Peninsula (Chevrizov 1978; Ferguson 1997) and two were recently described in two southern Japanese Islands (Sendra et al. 2018). In Haplocampa, the five described species have been found within the soil at high-altitude localities. However, Ferguson (1981) suggested no less than eight undescribed species of Haplocampa from caves located in the Ozarks karstic region at the centre of North America and in the small volcanic and karstic areas of the Pacific coast. Ferguson (1981) remarked that in the case of Haplocampa, Eumesocampa and Tricampa Silvestri, 1933, ‘All three have epigean members at high latitudes or high altitudes in the western mountains, and cavernicolous members at lower latitudes and altitudes’. Nevertheless, Haplocampa wagnelli sp. n. is an exception to Ferguson’s comment (1981).

H. wagnelli sp. n. is a slight troglomorphic species and several features show this. Among these features are the very slight elongation of antennae, 0.5× to 0.7× the body length, with moniliform antennomeres, and the cerci 0.5× to 0.6× the body length. In troglomorphic campodeids, the length of antennae can double that of the body, with always more than 30 antennomeres, and the cerci can be two or three times the body length; there is also an increase in the number of the cercal articles. Furthermore, troglobitic campodeids have elongated legs, reaching the end of the abdomen and a slim body. All of these features are trademarks of highly adapted subterranean campodeid species (Sendra et al. 2017). However, in H. wagnelli sp. n., its legs reach the VI to VIII abdominal segment and the body is not slender. But the trogloniont condition of H. wagnelli is supported by the presence of 32 antennomeres, five complex, multiperforated, folded-spiral sensilla, and a high number (14–16) of thick gouge sensilla. These are solid evidences for subterranean adaptations. For these reasons, H. wagnelli sp. n. can be considered to be a slightly troglobiont species adapted to live in caves and other subterranean spaces, but it could probably also be found in soil habitats. This is the case for the soil-dwelling species H. drakei, with 26 antennomeres, found nearby in Banff National Park, Alberta, Canada (Silvestri 1933).

H. wagnelli sp. n. may represent one of the most northerly cave-dwelling adapted dipluran species being found at 49° latitude north. The most northerly cave-dwelling species is Litocampa hubarti Bareth, 1999, found in Grotte Lyell (Liège, Belgium) at 50° latitude north, and like H. wagnelli sp. n., it shows humble morphological subterranean-adapted features (Bareth 1999). Both species are geographically in the limits of the Last Glacial Maximum. Furthermore, in the case of H. wagnelli sp. n., its current distributional area was under the Cordilleran Ice Sheet the Late Wisconsinian North American ice sheet complex, during the Last Glacial Maximum 18 ka BP (Dyke 2004). Several hundred years later the retreat of the glacial ice occurred, and as a consequence, the subterranean habitat colonisation of Vancouver Island was possible.