The exact page text, read by your browser's voice.
The town series, of record: 1910, 227 (corroborated by Bowen's 1917 table); 1920, 333; 1930, 522; 1940, 567; 1950, 736; 1960, 757 — the pre-2010 peak; 1970, 712; 1980, 629; 1990, 655; 2000, 567; 2010, 786; 2020, 879 (PL 94-171). The present estimate is the today article's ACS line (735, 2020–2024), printed there with its caveat. The anomalies are worked to documented size. Howard's 350 for 1900 against the census's 227 for 1910: three candidate explanations print at their strengths — different universes (Howard counted a village and its fringe; the census counted a place whose bounds incorporation fixed only in 1902), real decline (no exodus, fire, or closure is documented for it), and plain estimate error (proved possible in both directions by Bowen's 275 against 333) — the first and third together explain the gap without inventing an event; the settler that could close it is the 1900 manuscript enumeration for the village cluster. The 2000→2010 swing (+38.6%, the series' largest gain, hard after 1990→2000's −13.4%, its largest decline) prints unexplained at documented strength: no annexation ordinance was reached — the town's 2011 Comprehensive Plan nonetheless records the Morningside annexation on the northwest side, carried in the neighbourhoods record (assembly conform, DELTA-F-1) — and no carrier ties housing completions to the jump — a named gap (settlers: the Census geographic-relationship files, where boundary change is a yes/no fact, and the town's annexation ordinances). The township series was not settled and is not borrowed:other counties' Union Townships, flagged as a trap rather than used. The series is a named gap with a precise settler: STATS Indiana's township census counts, the St. Joseph County table read directly, reconciling to the 3,478 anchor.