An open reproduction licence has not been established. Nephos’s visual reading uses a captioned press reproduction, linked below.
A study against the weather record
Margery Austen Ryerson
Commercial Street, Provincetown
Nephos reads a sunlit street in a leafy season. What does the weather record reveal about the summer around it?
Regional light recursPainted day unknown
The Nephos method
How this study was read
The method in fullWhat is written
PAAM dates the panel to 1911; Quinn dates Ryerson’s first Cape Cod summer to that year. A contemporary report fixes the school session, not her attendance dates. Nephos reads the press reproduction as a sunlit street in a leafy season.
Date made: 1911
PAAM, object 3282.058.Pa25In 1911, Ryerson began spending summers on Cape Cod.
Karen E. Quinn, Margery Ryerson: Art Is Contagious (2025), p. 11Painting from landscape, still life, and the figure—indoors and out.
U.S. Bureau of Education, Summer schools in 1911, p. 1141with every one of the crew only too thankful to get home alive.
Daily Kennebec Journal, 31 July 1911, p. 1; Provincetown dispatch of July 30
Records of those days
- Provincetown daily recordNOAA USC00196676, about 5.2 km from Town Hall. Rain flags retained; presumed zeroes are unknown, not observed dry days.
- Blue Hill daily cloud and monthly sunshineSummit observations near Boston, about 78.5 km from Provincetown Town Hall. Printed daily 8 a.m. cloud and monthly Campbell–Stokes sunshine; a separate regional record.
- Then and now at Blue Hill1911 daily temperatures and NOAA 1991–2020 monthly normals for USC00190736. No station pooling or Provincetown successor splice.
Not found or not yet searched
- The painted day and Ryerson’s exact arrival/departure remain unknown.
- No verified local cloud series or 1911 Advocate weather passage was recovered; the checked online newspaper year list omits 1911.
- Open reproduction rights and a work-specific account of alterations remain unresolved.
Each claim tested
- How often did low-cloud mornings recur in the regional record during the two full school months? 27 of 62 mornings had at most two tenths cloud at Blue Hill. Bright conditions recur in both months; the painting cannot be assigned to any of them.
- Does a sunnier month mean no rain? Blue Hill recorded 62% of possible sunshine in July and 53% in August. Provincetown has 12 quantified rain entries and 2 traces in those two months; these are separate places and different measures.
- Can this identify a painted day or a heatwave scene? No. Cast shadows support direct light; they do not measure heat, and the regional cloud record cannot date this panel.
Verdict
A sunny street belongs within a changeable summer. Regional light records make the sunlit reading plausible without turning it into a dated weather report.
- Partially agreesRecurring opportunities for direct light in the regional school-season record The defined proxy occurs on 27/62 mornings; monthly sunshine supplies a separate measure at the same observatory.
- UnknownWeather agreement at the painted street and hour The scene’s date is unknown and Blue Hill is regional.
What this cannot tell you. July–August are the two full months within the school session of July 1–September 1, not documented execution dates. An 8 a.m. low-cloud reading is not a whole clear day. Sunshine duration is measured at Blue Hill, not at the street. The 1991–2020 comparison is descriptive, not a climate trend estimated from this painting.
Step 1 · Look at the picture
Sunlight, colour and a leafy street
Nephos’s reading of the picture, not a museum statement: a sunlit, apparently dry street in a leafy season, plausibly summer. Pale façades stand against broad blue-violet cast shadows across the road. Blue sky appears above the roofs; green foliage, pale clothing and brimmed hats support a warm-season association. The road reads as matte rather than puddled or reflective, a weaker clue than the shadows. The small sky patch does not establish a cloudless day, and the clothing cannot tell us the temperature.
We inspected the captioned reproduction in a 2022 ptownie feature on the Van Dereck collection. Its artist, title, date, medium and dimensions match PAAM’s record: Margery Austen Ryerson, Commercial Street, Provincetown, 1911, oil on panel, 12 × 15 inches, object 3282.058.Pa25. This is a press image, not an accession-linked museum master. An open reproduction licence has not been established; view it at the publisher.
Karen E. Quinn’s biography states: “In 1911, Ryerson began spending summers on Cape Cod.” A contemporary U.S. Bureau of Education report lists Hawthorne’s Cape Cod School of Art session as July 1–September 1, 1911. It describes the instruction as “Painting from landscape, still life, and the figure—indoors and out.” The table names Hawthorne as director, distinguishing his school from Webster’s nearby school.
The comparison uses July–August, the two complete months within that session. June appears only as wider summer context. These are school dates, not Ryerson’s established arrival and departure or this panel’s execution dates. Her first Cape Cod summer is documented to the year; an exact arrival day has not been found. PAAM’s catalogue year remains the object’s documentary date. Its record notes a signature on the back without transcribing it.
Step 2 · Test the sunlit reading
Bright mornings recur in both school months
The cast shadows make direct sunlight the useful question. How often did low-cloud mornings recur during the school months? We count the Blue Hill summit’s 8 a.m. cloud readings at or below two tenths of the sky. That is a stated proxy for opportunities for direct light, not a claim that a small cloud could not cover the sun. It also misses sunny breaks later in the day. The evening readings are retained in the table but do not enter the test.
27 of 62 mornings meet that rule: 16 in July and 11 in August. Tightening the rule to zero cloud still leaves 17 mornings across the two months. The Campbell–Stokes recorder provides a different measure at the same observatory: 278.0 hours of bright sunshine in July (62% of possible) and 219.7 hours in August (53%). This establishes recurring regional brightness rather than an unrelieved grey season.
The separation matters. Provincetown’s historical station is about 5.2 km from Town Hall on Commercial Street; Blue Hill, near Boston, is about 78.5 km away. Neither is the painting’s identified viewpoint. This Provincetown record is not a Highland Light record. The test supports the seasonal plausibility of our reading; it cannot verify sunshine on Commercial Street at the painted hour.
Every plotted day, with the retained evening cloud reading
| Date | Blue Hill cloud 8 a.m. | Blue Hill cloud 8 p.m. | Provincetown high | Blue Hill high | Provincetown rain |
|---|---|---|---|---|---|
| 1 July | 0 | 3 | 26.7 | 30.0 | Unknown (P) |
| 2 July | 9 | 1 | 29.4 | 34.4 | Unknown (P) |
| 3 July | 7 | 6 | 29.4 | 37.2 | Unknown (P) |
| 4 July | 1 | 6 | 26.7 | 37.2 | Unknown (P) |
| 5 July | 0 | 8 | 28.9 | 36.7 | Unknown (P) |
| 6 July | 0 | 10 | 28.3 | 36.7 | Unknown (P) |
| 7 July | 8 | 0 | 23.3 | 24.4 | 9.4 |
| 8 July | 0 | 0 | 26.1 | 28.3 | Unknown (P) |
| 9 July | 1 | 4 | 26.7 | 32.8 | Unknown (P) |
| 10 July | 0 | 3 | 33.9 | 37.2 | Unknown (P) |
| 11 July | 0 | 2 | 32.2 | 36.7 | Unknown (P) |
| 12 July | 3 | 8 | 32.2 | 34.4 | Unknown (P) |
| 13 July | 0 | 2 | 33.9 | 30.6 | Unknown (P) |
| 14 July | 10 | 6 | 26.7 | 25.6 | Unknown (P) |
| 15 July | 0 | 7 | 26.7 | 28.9 | Unknown (P) |
| 16 July | 7 | 2 | 26.1 | 30.0 | Unknown (P) |
| 17 July | 9 | 10 | 24.4 | 24.4 | Unknown (P) |
| 18 July | 2 | 7 | 26.7 | 29.4 | Unknown (P) |
| 19 July | 0 | 3 | 26.7 | 27.8 | Unknown (P) |
| 20 July | 9 | 9 | 26.7 | 27.2 | Trace |
| 21 July | 9 | 8 | 23.9 | 24.4 | Trace |
| 22 July | 9 | 0 | 26.7 | 28.3 | Unknown (P) |
| 23 July | 0 | 10 | 27.8 | 27.8 | Unknown (P) |
| 24 July | 10 | 10 | 24.4 | 23.9 | 16.5 |
| 25 July | 0 | 0 | 24.4 | 24.4 | Unknown (P) |
| 26 July | 1 | 1 | 23.9 | 25.6 | Unknown (P) |
| 27 July | 6 | 9 | 22.8 | 24.4 | Unknown (P) |
| 28 July | 10 | 10 | 20.6 | 18.3 | 57.4 |
| 29 July | 8 | 9 | 23.9 | 23.9 | Unknown (P) |
| 30 July | 10 | 7 | 25.0 | 24.4 | 1.3 |
| 31 July | 1 | 6 | 29.4 | 30.6 | Unknown (P) |
| 1 August | 2 | 10 | 25.6 | 26.1 | Unknown (P) |
| 2 August | 10 | 8 | 21.1 | 17.8 | Unknown (P) |
| 3 August | 10 | 0 | 20.0 | 22.2 | Unknown (P) |
| 4 August | 7 | 0 | 22.2 | 23.9 | Unknown (P) |
| 5 August | 9 | 10 | 21.1 | 22.8 | Unknown (P) |
| 6 August | 10 | 10 | 21.1 | 21.7 | Unknown (P) |
| 7 August | 10 | 0 | 20.6 | 25.0 | Unknown (P) |
| 8 August | 10 | 2 | 21.1 | 25.6 | Unknown (P) |
| 9 August | 10 | 7 | 22.2 | 24.4 | Unknown (P) |
| 10 August | 0 | 4 | 27.2 | 32.2 | Unknown (P) |
| 11 August | 7 | 10 | 27.8 | 26.7 | Unknown (P) |
| 12 August | 1 | 6 | 25.6 | 25.6 | Unknown (P) |
| 13 August | 2 | 1 | 25.6 | 27.8 | Unknown (P) |
| 14 August | 0 | 0 | 26.7 | 30.6 | Unknown (P) |
| 15 August | 3 | 10 | 27.8 | 31.1 | Unknown (P) |
| 16 August | 0 | 0 | 26.7 | 25.6 | 10.9 |
| 17 August | 9 | 1 | 26.1 | 28.3 | Unknown (P) |
| 18 August | 3 | 10 | 26.7 | 28.3 | Unknown (P) |
| 19 August | 2 | 0 | 23.9 | 22.2 | Unknown (P) |
| 20 August | 0 | 0 | 24.4 | 23.3 | Unknown (P) |
| 21 August | 0 | 0 | 25.6 | 25.6 | Unknown (P) |
| 22 August | 0 | 0 | 23.3 | 24.4 | Unknown (P) |
| 23 August | 2 | 10 | 26.7 | 30.6 | Unknown (P) |
| 24 August | 3 | 10 | 22.2 | 21.7 | Unknown (P) |
| 25 August | 10 | 10 | 18.3 | 16.1 | 16.0 |
| 26 August | 10 | 10 | 20.6 | 21.7 | 9.1 |
| 27 August | 10 | 10 | 17.8 | 16.7 | 27.9 |
| 28 August | 10 | 10 | 26.1 | 25.6 | 1.3 |
| 29 August | 10 | 9 | 26.1 | 22.8 | 6.6 |
| 30 August | 10 | 10 | 17.2 | 15.0 | 3.0 |
| 31 August | 10 | 10 | 15.6 | 13.9 | 31.0 |
The original definitions keep the measures distinct. A published “clear day” uses hourly cloud observations from 7 a.m. to 8 p.m.; our morning count does not reproduce that daily classification. Bright sunshine records sufficient direct light to burn a Campbell–Stokes card. The historical calculation of possible sunshine subtracts thirty minutes per day for the low sun near the horizon.
Step 3 · Hear the season’s witnesses
The same summer brought a destructive gale
Provincetown’s largest measured summer rain entry is 57.4 mm on 28 July. A Provincetown dispatch dated July 30, printed in the Daily Kennebec Journal on July 31, reports four local fishermen lost and damaged vessels returning after the previous Friday’s gale. It describes one returning crew as:
“with every one of the crew only too thankful to get home alive.”
Daily Kennebec Journal, “Four Provincetown Men Lost,” July 31, 1911, p. 1
The report concerns vessels and offshore losses; it is not an eyewitness account of this street. Its Friday is July 28. The gauge and the dispatch put a substantial rain entry and a reported coastal disaster on the same date, without reconstructing hourly conditions or claiming that Ryerson witnessed either. We retain the newspaper’s report without adopting its storm terminology as a modern meteorological classification.
Across July–August, the local file contains 12 positive rain entries and 2 traces. Its other 48 entries carry NOAA’s missing-presumed-zero flag. They stay unknown: no observed dry-day count or complete local rain total is claimed. A daily rain entry would not by itself rule out a sunlit interval earlier or later.
The broader June–August local temperature record is complete: mean daily highs 23.9 °C, lows 15.3 °C, and a maximum of 33.9 °C on 10 July and 13 July. Those hot days remain context, not candidate dates selected for the painting.
The Provincetown Public Library’s digitized newspaper year list omits 1911. No 1911 Provincetown Advocate weather note or Ryerson weather letter was verified in this search. That is a retrieval gap, not a claim that none survives. The school report and the Provincetown-datelined dispatch are the contemporary witnesses used here.
Step 4 · Then and now; what the painting selects
A hot July did not make every summer day hot
A modern comparison is possible at Blue Hill alone. The 1911 daily series and NOAA’s 1991–2020 normals share station identifier USC00190736. July’s mean daily high was 29.2 °C, 1.6 °C above the modern July normal; August’s was 24.0 °C, 2.7 °C below its modern normal. A single summer could alternate between a hot month and a cooler one. The painting’s brightness cannot tell us which it represents.
| Month | 1911 low / high | 1991–2020 normal low / high | High departure, °C |
|---|---|---|---|
| June | 12.6 / 22.1 | 14.2 / 24.1 | -2.0 |
| July | 17.8 / 29.2 | 17.7 / 27.6 | +1.6 |
| August | 14.5 / 24.0 | 17.0 / 26.8 | -2.7 |
The normal is a processed climate product, whereas the historical values here retain the daily archive’s quality flags. This is a descriptive comparison, not a homogenized estimate of climate change. There is no modern normal for the painted street in this study, and no local successor station has been joined to the old Provincetown file.
What Ryerson chose: in Nephos’s reading, the broad pale wall and diagonal bands of shadow make colour contrast structure the busy street; the small figures and touches of red and blue punctuate it. Quinn describes Hawthorne’s emphasis on colour relations and Ryerson’s response in other Cape Cod works. That is useful teaching context, not evidence that this panel was a class exercise or that Ryerson altered its weather. No matched sketch, letter or technical examination establishes what she changed.
The finding: the picture presents a bright interval within a season that had recurring regional sunshine, local rain and a reported coastal gale. Its light is worth reading as evidence; its exact day remains unknown. A photograph or transcription of the reverse, the catalogue’s basis for 1911 and work-linked correspondence could make the next test more specific.
Source scans, manually checked transcriptions and derived values are identified separately in the receipts below. Hash checks establish reproducibility, not independent validation of the instruments or the visual interpretation.
Sources
Every source on this page
Each file was downloaded and given a fingerprint (its sha256). The page is rebuilt from these files and will not build if a fingerprint changes, so every number above comes from them rather than being typed in.
- PAAM, Commercial Street, Provincetown, object 3282.058.Pa25
sha256 372792a9ac2abc885a82509314796f1d92ea7869123a3d45170fe60b90377ab7 - PAAM, Van Dereck Collection: Part II, checklist
sha256 923537a796daf1a80c18b6477a4bfdbce38af85258d6685cd8134c5297c7e692 - Karen E. Quinn, Margery Ryerson: Art Is Contagious (NYSM, 2025)
sha256 f4a4776bd93d0c6a70e186508b08aee846465cffea9c4e179ce77848b9fad01d - Text extraction of Quinn, Margery Ryerson: Art Is Contagious
sha256 c751c7963561a8c4b3915590432231be762c51890a941c399da6f81e0e4f2187 - NOAA NCEI, GHCN-Daily USC00196676, Provincetown 3 NW
sha256 edfa7779bc3a9cdaa7467b5e0d135ef3be79febe8af1b5b613fc452fbfe4496f - NOAA NCEI, GHCN-Daily format and flag definitions
sha256 0e4e6f0e6947cf1add884c328a01760f31c56ef5fb86b85ed57dd5dc4c4aeaa6 - NOAA NCEI, HOMR station history queried for 1 July 1911
sha256 86eb9542631d6e89001eadd8c5199976f312a53b4449b6902ccf30f11c3734e9 - OpenStreetMap, Provincetown Town Hall, 260 Commercial Street
sha256 f50a78d79dc7a0c156988a86966bcd5ca72c5c7be58307e82c4b50893b72feab - Provincetown Public Library digital newspaper archive
sha256 1b68020424ef1c5d613b252ebad074239511107daf744ca881907c5af699ffbd - ptownie, Napi Van Dereck – The Way It Used To Was
sha256 51f25b4e212a2ae5c253a41a7f88dcb865481c75c205dda829157b0d584dd4c4 - Ryerson press reproduction, as captioned by ptownie
sha256 b311f86b1603116b343f718a7cd581d44bf64006ebde334f0b0557a8f7c5f1e6 - U.S. Bureau of Education, Summer schools in 1911
sha256 58b3179a01a18326193101bc7f2e4926caba9d47add3acf1ee3743169408858d - NOAA NCEI, GHCN-Daily USC00190736, Blue Hill COOP
sha256 ce49b9a5d0e91d2f0193f4621f7df6438fe2089993706827d36b86df2235dd07 - NOAA NCEI, Blue Hill history queried for 1911
sha256 2c4665dcd00957996602b3b967a77b379a886520947ee2d553a785fd99d0547a - Blue Hill observations, Harvard Annals 73, part II (1915), p. 91
sha256 0af8de0068ac1198de0b0206b7085859ca6ad2f609ead3dbc54f551576715b53 - Blue Hill observations, Harvard Annals 73, part II (1915), p. 93
sha256 75b4c54b10fd0c785a05f5f25a427b8c34e4688d2d76f0535816106c557552d8 - Blue Hill observations, Harvard Annals 73, part II (1915), p. 95
sha256 04028d6f4fb6e1e67fd61b80e5e2e94fff4724cb354f8b7262001ea972c3e73a - Blue Hill observations, Harvard Annals 73, part II (1915), p. 99
sha256 4d489150ba489e9b1d38d66fea7ec1673d791d208f6a21cf46214104cae17528 - Blue Hill observations, Harvard Annals 73, part II (1915), p. 107
sha256 e446d1af1252893be2fac247da40efe0349b5c9b34c3350b8fca532380c235e1 - Blue Hill observations, Harvard Annals 73, part II (1915), p. 108
sha256 03460997102132cdd448ab54c650d2b03f54d3382e040638109f81530d5e0703 - Blue Hill observations, Harvard Annals 73, part II (1915), p. 109
sha256 466496061840c0c632a244cf3dd374927c659461014d188dfd84952569678efd - Blue Hill observations, Harvard Annals 73, part II (1915), p. 110
sha256 9055062d6ad78f177ca5090700e505e9a86b9ad2427b642d59f117addd536534 - Blue Hill observations, Harvard Annals 73, part II (1915), p. 114
sha256 457ed02a27e46ed00f71d2fe5ca34b6b2a9240ac091781171e18d2f1e45849ff - Blue Hill observations, Harvard Annals 73, part II (1915), p. 115
sha256 b4855414ef6ff97573a5cce2e325fd1f497591d09df45116c36a975ca1c06783 - Blue Hill observations, Harvard Annals 73, part II (1915), p. 116
sha256 088ce929718e959a799433ba5c51a08850323bb6f3a996ea5154fd74703e2d73 - NOAA, Blue Hill COOP 1991–2020 monthly normals
sha256 539bebd47a069b90ecb9084e5d7356f93dab8b09485c34389a48623565a3dd5c - NOAA, 1991–2020 monthly normals documentation
sha256 b12c9aee6770f84b5fa902740413bffca5cbf9a83bbebcf7e73306f46168ffc6 - Daily Kennebec Journal, 31 July 1911, p. 1
sha256 5535a91cb304656ab76b5d075cb7ce53f2baf323ed47af63e6e5b9b35b6bd71a - LOC text-service response for Daily Kennebec Journal, 31 July 1911
sha256 4e149fa9811b087eed01650b7522313593c72458992f0eeb020f53797b97a1d1 - Nephos, scan-checked Ryerson revision transcription
sha256 a01aacb32d7b7f6a3d8ccbe2f45c0980d25edbf2b002cca2a29dbc93b50e8263