Flaws in the sources

Every parent-child link on this site is backed by a primary source. Inside those primary sources, flaws the source carries about itself sometimes turn up.

They are not corrected. They are quoted as they stand, with a note. This page says why, and lists the ones found.

Why they are not corrected

Where a number is mistyped, the right number can be read off another field. Correcting it would sometimes read better.

They are left alone because the moment a quotation is corrected it stops being a quotation.

The only thing this site is worth is being able to say "the source reads thus". Let one line of "the source reads thus, but we have fixed it" into the mix and the reader can no longer tell which lines are the source and which have been touched.

So places known to be wrong are quoted wrong, with a note beside them.

Wrong on the paper itself

Mistakes by whoever wrote or set the report.

Document	As written	Should be	How we know
Hatsunishiki (Ou No.224)	Ou No.244	Ou No.224	The same document's regional number and ineCode are 224
Tedoriwase (Etsunan No.6)	Etsunan No.5	Etsunan No.6	In the conclusion of the Ishikawa section, against the document's own variety
Paddy Rice Norin No.42 (Hokkai No.103)	Paddy Rice Hokuriku No.103	Hokkai No.103	The breeding site is the Kamikawa Branch in Hokkaido
Mineasahi (Chubu No.17)	奨助 variety	奨励 — recommended variety	Elsewhere the same document writes it correctly

All four are mistakes in a number or a word, not in a description of character or of a year. Where a source goes wrong is not evenly distributed.

Lost on the way from paper to screen

These are not the report's mistakes but damage from digitisation.

"？先" — the same part written four ways

Awn and ？ apiculus colour　Awnless, and the ？ apiculus is red. (Yamaseshirazu)

The colour of the tip of the husk. Across the database it is written four ways.

Spelling	Documents
ふ先	Natsuminori (Tohoku No.58)
桴先	Wakakusa (Etsunan No.39)
？先	Yamaseshirazu (Tohoku No.63), Wakasa (Etsunan No.15), Yamahonami (Etsunan No.18)

The "？" is not one accident but a habit of the digitisation. The paper probably used a character such as 稃 that could not be rendered; since there is no telling which, it is quoted as it stands.

Characters whose shape has broken

Mineasahi (Chubu No.17) carries four.

As written	Should be
出徳期	出穂期 — heading date
極数は中位	穂数 — panicle number
カラバ工	カラバエ — stem maggot
奨助品種	奨励品種 — recommended variety

The newer documents appear to carry more of this. The old ones were set in type; the new ones were read straight off paper. Mineasahi's document is from 1980, the newest on this site.

Paddy Rice Norin No.19's "the Soya 船 districts" also appears to have broken, but since there is no telling how, the intended form is not written here.

Things simply missing

Document	What is missing
Paddy Rice Fujisaka No.4	"in [blank] 1948" — the month
Yamaseshirazu	The punctuation in "grain set is dense and shattering hard the grain is somewhat large"
Harukaze	The attached file itself (file name: Kanto No.74 districts of adoption), which the districts are left to

Harukaze's attachment is not in the database. When a separate sheet or a figure drops out, the text that points to it stays behind.

It is not only paper documents that lose things. On the Masshigura page in the variety database of the Aomori Prefectural Industrial Technology Research Center's Agriculture and Forestry Research Institute, there is a heading reading "The origin of Masshigura" — and nothing under it. On the breeding institute's own page, only the frame of the origin section is left (checked 8 October 2026). We have still not found a source that gives this variety's year of crossing.

Where sources disagree with each other

This is not damage but two sources saying different things. Neither is picked; it is recorded as a dispute.

About	The disagreement
Koshihikari's year of cross	Four NARO historical documents say 1945; Niigata Prefecture and the JIII say 1944 (the text takes 1944)
Kaganishiki's culm	Its own document: "stronger than Norin No.1, strong culm". Tedoriwase's, from the same prefecture three years later: "slightly stronger, culm not strong, lodges easily"
Kurenaimochi's parent	The historical document says "Hokuyoku"; Koshihikari's listing of children says "Houyoku"
Tomohonami's parent	Its document names Mineasahi as mother and Senshō as father and never mentions Koshihikari, against the cross combination in the listing (no page is made until it resolves)

Some of the damage is ours

Everything above is damage in the sources. There is damage on our side too.

In the variety registration database, the "origin" sentence always ends with "… non-glutinous paddy rice." Of the 53 such passages quoted on this site, 19 stopped before they got there.

…impaired-type cold tolerance and leaf bl (Hamakowase, as the quotation stood while it was cut) …an estimated blast-resistance genotype of Pi- (Omachikane, the same)

They broke mid-word, so that was not how the source reads. They were cut when we captured them. On 8 October 2026 all nineteen were taken again from the originals and replaced. The English and traditional-Chinese versions had rounded the cut off into a sentence, so those were rewritten too.

While we were at it, all 53 passages were compared against the originals by machine. The other 34 did not differ by a single character, but 30 had dropped the full stop at the end of the sentence. That was added.

The stray space inside the Okuhikari quotation turns out not to be ours: the source writes it that way. The same sentence has a second one. Both stay.

There was one more fault on our side. On the spelling of Hananomai, this site said that NARO's list wrote it with a trailing "i" and the variety registration without. The two spellings were attributed to the wrong institutions. The registration's own name field reads "はなの舞い"; the form without the "i" is what the same database writes in the origin sentence for Kako. The spelling does not split along institutions — both forms sit inside the same source.

The third fault on our side was not reading a document to the end. On the breeding record for Tomohonami (Chubu No.125), this site said the record never mentions Koshihikari. It does. A later paragraph states that Koshihikari was crossed twice onto "Inakei IL946" in 2001-2002. We had read only the first sentence and called it a disagreement. Reading the whole document on 8 October 2026 showed there was none.

Damage in a source we leave as it is; damage on our side we repair.

How we noticed is worth recording too. While fixing pages where emphasis marks were showing in the body text, we compared the endings of the quotations by machine. Reading them one at a time, we had not noticed. The Hananomai fault surfaced because taking the cut quotations again meant opening the name field, where the name did not match our own page.

For the reader

That a source carries flaws does not make it unusable. We can say where the flaws are only because other fields and other documents could be set against it.

This site lists every source so that the reader can make the same comparison again, by hand.