Redundant link removal raised remote voltage in passive models of two partial adult worm maps

We tested remote voltage transfer after removal of one electrical link in two partial adult C. elegans maps. All tested links could be removed without graph disconnection. A passive model had uniform membrane leak and contact-count conductance. At leak equal to 0.1 times median intact weighted degree, more than 10 percent fixed-current voltage increases occurred in 15,702 of 4,781,520 link and ordered source-target combinations in adult 7 (0.328 percent) and 5,460 of 5,730,354 in adult 8 (0.0953 percent). Neither source nor target touched the removed link. Fixed-source-voltage increases occurred in 0.256 and 0.0713 percent. Increases persisted with unit link weights and after exclusion of weak baseline responses. Their shares changed with leak. A standard inverse identity gives the exact change sign from intact endpoint voltage gradients. Direct full solves confirmed all primary threshold counts. Two incomplete maps and assumed passive parameters prevent claims about live-animal responses or reliable computation. Code, matrices, edge counts, hashes, retrieval steps and a figure support reproduction.

C. elegans · gap junction · membrane leak · passive transfer · link removal · circuit robustness