1. Amynthas agrestis (Goto & Hatai, 1899)

(Figures 1 A, 7; Table 1)

Perichaeta agrestis Goto & Hatai, 1899: 17, 24.

Pheretima agrestis — Howell 1939: 231. Gates 1953: 5; 1954: 224; 1958: 1, 31; 1963: 11; 1982: 38.

Amynthas agrestis — Sims & Easton 1972: 235. Reynolds 1978: 119, 127; 2010: 143; 2011: 269. Reynolds & Wetzel 2004: 88; 2008: 179. Blakemore 2010a: 429; 2013b: 56, 57.

Metaphire agrestis — Blakemore 2003: 7, 28.

Data sources. Goto & Hatai (1899); Gates (1953, 1954, 1982); Blakemore (2010a, 2013b); this study (USNM 1421431).

Diagnosis. Size 70–160 mm by 5–8 mm. Segment numbers 63–110. Color of live specimens red. Male pores usually absent; when present, small, transversely slit-like. Post-clitellar genital markings usually absent; when present, single, large circular pad, pre-setal on XVIII, just median to male pores, with a concave center surrounded by a narrow but distinct, raised rim, reaching posteriorly slightly behind the setal line on XVIII and anteriorly to the setal line on XVII. Spermathecal pores three pairs in 5/6/7/8 or variously missing. Pre-clitellar genital markings present or absent; when present, ventral, areas of slight epidermal modification on VII and/or VIII, occasionally on VI and IX, unpaired and median or symmetrically paired, forming setal gaps, epidermis finely wrinkled or crosshatched, sometimes darker in color in live specimens. Female pore single in XIV. First dorsal pore 12/13. Spermathecae present or absent; when fully present, three pairs in VI–VIII, duct shorter than ampulla; diverticulum longer than duct and ampulla combined. Prostate glands present or absent; when present, extending through some or all of XVI–XXIII, ducts in XVIII. Intestinal caeca paired in XXVII, manicate.

Amynthas agrestis Amynthas tokioensis Metaphire hilgendorfi Prostate glanđs Usually absent Present or absent Usually absent

Intestinal caeca Manicate Manicate Manicate

Length by wiđth. While the ranges of size are similar among the three species, A. tokioensis is generally consiđeređ a small species anđ is smaller than the other two species.

Remarks. Amynthas agrestis has been frequently reported in the continental US, and is one of the two pheretimoid species recorded in Canada. The first record of this species in the continental US was in 1939 from the Homewood campus of the Johns Hopkins University, Baltimore, Maryland (Howell 1939; Gates 1954; 1982). That record was the second of A. agrestis outside Japan, where the species is native. It has recently been confirmed to be abundant in Baltimore and has been observed to co-occur frequently with M. hilgendorfi and/or A. tokioensis in Maryland, Connecticut, Vermont, New Hampshire, and Wisconsin (C.- H. Chang, personal observation), two species morphologically similar to A. agrestis (Table 1). This latter observation suggests the high possibility of misidentification. Reproduction of A. agrestis in North America is parthenogenetic. In field conditions, A. agrestis, an annual species, overwinters only as cocoons and the adults reproduce in summer and die by the end of fall (Callaham et al. 2003; Richardson et al. 2009; Görres et al. 2014). However, in the laboratory, its adults can survive through November–February (Snyder et al. 2013; Ikeda et al. 2015). Amynthas agrestis is epi-endogeic, and its successful invasion in US forests has been attributed to dietary flexibility (Zhang et al. 2010). It has been known to compete with native millipedes in southeastern US for food resources, particularly the fragmented, partially decomposed leaf litter (Snyder et al. 2011; 2013). Current practice of using commercial mulch for horticulture and landscaping may help spreading of this invasive species (Belliturk et al. 2015).

For years, what the male pores and the associated genital markings of A. agrestis look like has remained an unanswered question until our study. Male pores are almost always absent in A. agrestis . They were rarely described, and have never been illustrated before. Blakemore (2010a) illustrated the male pores of a Japanese specimen that was believed to be A. agrestis . However, later he questioned his earlier conclusion and noted that the previously reported specimen may belong to a different species (Blakemore 2014). Gates (1982) apparently saw one A. agrestis specimen with a male pore on one side, as his detailed description of the male pore and genital marking matches our Figure 5 A. However, his observation was confounded by having two specimens with different male pores and he even questioned whether he had two species.