The case
Cicadas carry two nutritional endosymbionts, Sulcia and Hodgkinia.
- Splitting (LIT-1563). In some Tettigades, Hodgkinia has split into two cytologically distinct species. Their gene loss and retention are almost perfectly complementary, so a three-member symbiosis became four. The authors suggest nonadaptive forces can build organismal complexity.
- Repeated (LIT-1586). A single ancestral lineage split at least six independent times in Tettigades over four million years, giving two to six lineages per host. Together the complexes encode nearly the ancestral gene set.
- Fragmentation (LIT-1612, LIT-1602). In Magicicada tredecim the genome fragmented into many chromosomes, together almost ten times the ancestral size. Every Magicicada species holds at least twenty circles, with little synteny between species and very different gene dosages.
- The host’s cost (LIT-1590). Rare but essential lineages risk being lost at transmission. Eggs still receive the full set, because hosts with more complex Hodgkinia transmit up to six times as many cells. The authors propose this as a cost of the fragmentation.
What it can show
- More parts, more dependence, no new function. The complexes do what one lineage did. Integration measured by dependence rose with no evident gain, which is A2’s “partly nonadaptive” dependence (QUESTION-277).
- Host compensation at transmission. The host’s response raises reproductive cohesion (R) to rescue a collective made fragile by constituent changes.
- Constituent change that costs the whole. This is the converse direction of CLAIM-060. What the constituents became is not what the collective needed.
Documented in works the record has not read. What it shows rests on their abstracts until they are read.
Note of 2026-10-27 (batch 2): the transmission paper read in full
LIT-1590 has been read in full (NOTE-1235, THEORY-1106). Corrections to “The host’s cost” and “What it can show”:
- The six-fold rise is one species. It compares M. septendecim (30,154 Hodgkinia cells per egg) with a single-lineage Tettigades (4,889). Only M. septendecim differs significantly. The six-lineage T. chilensis (about 12,000) does not differ from the single-lineage species. The study has four species and 5–6 eggs each, with no phylogenetic control.
- “Eggs still receive the full set, because …” is inference. The full set was tracked with one or two marker genes per species, and the authors grant that rare losses could escape detection. The “because”, and the order (splitting before titre), are the authors’ inference.
- The hundredfold abundance range comes from the authors’ earlier metagenomes (114-fold in rpoB), not from this paper.
- Circles are not shown to be cells. In Magicicada, genomic circles are not shown to be separate cell lineages. The simulation assumes they are, and the authors suspect some circles share cells.
- The host maintains R; it does not raise it. It keeps co-transmission complete as the parts multiply, by sending more Hodgkinia (not more Sulcia) through an unchanged route. The authors speculate that the larger transmitted population weakens bottlenecks and “decrease[s] the level of control that the host can exert”, which could feed further splitting. If so, keeping R costs C and S.
- Composition is inherited. Mothers pass their own lineage proportions to their eggs, and egg nests differ significantly, so the make-up of the symbiont community is inherited at the host level.