The Lv (吕) Surname: A Prime Minister’s Pedigree and No Ancestral Village

Seven Million People, No Village

The Lv surname — 吕 in Chinese, Lǚ in pinyin, Lv typed on a keyboard that has no diacritics — is a top-forty surname: about 7.0 million bearers, 0.44 percent of the population, fortieth on 2018 figures. In Taishan it is nowhere. The county gazetteer 《台山县志》 (1998) counts villages by surname — 黄 291, 陈 232, 李 222, 伍 170 — and 吕 is recorded in none of them.

Four characters have to be separated before anything else. 吕 (Lǚ) is not 陆 (Lù, land), the surname that owns most searches for “Lu surname”, which has its own article. It is not 卢 (Lú), the Fanyang Lu, near enough in sound to be interchanged in genealogies and on remittance envelopes. And it is not 雷 (Léi, thunder), the other Lui, which has an article too.

The pedigree behind the character is old and unusually well lit. 伯夷 (Bóyí), a descendant of the 四岳 (Sìyuè, the Four Peaks), was enfeoffed at 吕 in the 虞夏之际 (Yú–Xià zhī jì), the transition between the two founding dynasties. 吕尚 (Lǚ Shàng) — 姜太公 (Jiāng Tàigōng), the strategist behind the Zhou conquest of the Shang — took his surname from that fief. 吕不韦 (Lǚ Bùwéi) ran Qin as chancellor, 吕雉 (Lǚ Zhì) ruled as Han empress,

Most 吕 families are told they came down 珠玑巷 (Zhūjī Xiàng), the lane in 南雄, a corridor norm well documented for other surnames. The roster for that Song flight, 罗贵’s thirty-three surnames and ninety-seven households, contains 卢 and 陆 and not 吕.

Hence the working picture. The Guangdong route ran by sea: 晋江 → 潮州胪岗 → 丰顺 → 海丰. 四邑 (Sìyì) is a documented absence with one line in a 1963 index, filed under a township that no longer exists and marked 待证 (dài zhèng, to be verified) — a household, not a village. And the family often cannot be found because the name is Lui, which is also 雷.

A State, a Code, and Two Ancestor Slots

The oldest recoverable sentence about this surname is not a pedigree but a grant. 《史记·齐太公世家》 records that the ancestor helped 禹 control the waters and 「虞夏之际封于吕」 — was enfeoffed at 吕 at the turn of Yu and Xia. A second clause does the real work: 吕尚 「本姓姜氏，从其封姓，故曰吕尚」, he was of the Jiang clan by birth and took his surname from his fief, which is why he is called Lü Shang.

So 姜 (Jiāng) is the 姓 (xìng), the clan name, and 吕 is the 氏 (shì), taken from the place: 姜姓吕氏. The Jiang article owns that grammatical system. What matters for a state-origin surname is that it keeps two ancestors, in separate slots, unmerged in the conventions examined here. 伯夷 is the 受姓祖 (shòuxìngzǔ), the ancestor from whom the surname derives; 吕尚 is the 精神始祖 (jīngshén shǐzǔ), the spiritual founder a hall is named for.

The surname’s first appearance in the canon is a code. 周穆王 appointed 吕侯 — also written 甫侯 (Fǔ Hóu) — as 司寇 (sīkòu, minister of justice), and the result is 《尚书·吕刑》: 「五刑之疑有赦，其审克之」, where a penalty is doubtful, pardon it; weigh the case carefully. That is the earliest canonical text cited here, a canonical attribution rather than a biography. The 吕/甫 equivalence rests on 《诗经·崧高》 and on a bronze interchange of the two graphs.

The ground is 董吕村 (Dǒnglǚ cūn), now 董营村, in 王村乡, 卧龙区, Nanyang, thirty 里 west of 宛城 (Wǎnchéng). It is a 文献地望 (wénxiàn dìwàng) — a documented location, not an excavated fact. Two sites stay in play: 董营村, and 镇平县侯集镇, said to have yielded artefacts, in a report that concedes the location 「尚待考证」. The 吕侯墓 became a protected site in 2016 and a reconstruction began in 2020; 吕方鼎 and 吕王鬲 are cited as corroborating bronzes, with no excavation report located. 吕耕’s 《吕姓溯源》 puts the first enfeoffment near 霍太山 in Shanxi and Nanyang second: a live alternative, adjudicated nowhere.

Dispersal came by defeat. 楚 absorbed 申 and 吕 around 688 BCE, though 《左传》 gives no explicit year and one strand dates the Shen campaign to 679; part of the population moved east to 新蔡 (Xīncài) as 东吕, whose end one tradition dates to 553, the accounts conflicting. Then 田氏代齐 in 386 BCE pushed the dispossessed Jiang house out of Shandong, so Shandong 吕 is a confluence of the old state’s remnant and the Qi ruling clan. No documented migration wave appears in the sources used here; and because this dispersal runs two centuries ahead of Qi’s, a pedigree that opens with the Tian usurpation dates its own origin too late.

吕洞宾 (Lǚ Dòngbīn), the immortal of the Eight, belongs to religious tradition, not to genealogy, and no reader descends from him.

The Plaque Is Not the Address

Almost every overseas 吕 hall hangs a 河东 (Hédōng) plaque, and readers take it for an address. 《读史方舆纪要》, volume forty-one, under 霍州 (Huòzhōu), says something else: 「州西三里，故吕乡，晋吕甥邑也」 — three li west of the prefectural city lies the old Lü township, the fief of Lü Sheng of Jin.

One major tradition therefore identifies 河东 吕 with a 姬姓 (Jī xìng, Ji-clan) fief of Jin held by 吕甥 (Lǚ Shēng), a minister of Duke Hui, near modern Huozhou in Shanxi; it should not be collapsed into the Jiang-clan state at Nanyang. A second runs through 魏锜, styled 吕锜, whose son 吕相 (Lǚ Xiàng) took Lü as a surname and delivered the 绝秦 (jué Qín) address, Jin’s severance from Qin, in 578 BCE. Genealogical accounts trace 河东 吕 to these two routes; they are traditions, not verified lineages, and neither touches Nanyang.

The consequence lands on precisely the branch whose hall name most overseas associations hang: for it, “pure Jiang” does not hold. A third stem enters from the north, when the Northern Wei emperor Xiaowen’s reforms turned 叱吕氏 (Chìlǚ shì) into 吕, as 《魏书·官氏志》 records.

河东’s seat was 安邑 (Ānyì), today 夏县 禹王城 — so 河东堂 names a site in Shanxi, not in Henan.

The codification came late. 河东, 东莱, 南阳 and 金华 were fixed as canonical 郡望 (jùnwàng, commandery-of-origin prestige labels) in the Tang 《元和姓纂》, not in Han or Wei times, and 晋江 appears in later lists as a prestige label rather than a commandery.

Hence the split that makes every plaque honest and useless at once: 金华 吕氏 lived for generations in 婺州 (Wùzhōu) while styling itself 东莱 吕氏, and a Minnan hall can carry 河东 as 郡望 and 理学 — Donglai, the sobriquet of 吕祖谦 — as its 堂号 (tánghào, hall name).

The couplet does the same work: 河东世泽，渭滨家声 — the grace of Hedong, the renown of the Wei bank — a commandery and a fishing story, no village at all. A Lingnan compilation gives a variant, 莆田世泽，渭水家声; that suggests a Putian-associated memory but does not by itself prove a route.

So the diagnostic rule, which is the tool the rest of this article uses: a branch’s route is read from its 字辈 (zìbèi, generation characters) and its 村名 (village name), never from the plaque.

What the Surname Actually Gave Away

The phrase “one house, three chancellors” — 一门三相 (yī mén sān xiàng) — fuses two unrelated houses. 吕蒙正 (Lǚ Méngzhèng) was a Luoyang man, first in the 977 palace examination and chancellor three times. The other house begins with 吕龟祥, posted to 寿州 that same year, 977, whose grandson 吕夷简 and great-grandson 吕公著 also reached the chancellorship. 《宋史》 is itself ambiguous about the link, and the tie it states is a Laizhou ancestral registration, not Luoyang. The fusion is the tell: a narrative that repeats three chancellors as one pedigree tells you which genealogy it read.

One line survives source criticism. 东莱 (Dōnglái) runs 吕夷简 → 吕公著 → 吕希哲 → 吕好问 → 吕本中 → 吕祖谦, three holders of the highest offices across seven generations, anchored by 吕祖谦’s 鹅湖之会 (Éhú zhī huì), the Goose Lake debate of 1175, and by his 《皇朝文鉴》 in 150 chapters. Among the lines examined here, only this one survives that test.

Around 熙宁九年 (1076) the four 吕 brothers of 蓝田 — the 蓝田四吕 — produced 《吕氏乡约》: 「德业相劝，过失相规，礼俗相交，患难相恤」. Encourage one another in work, correct one another’s faults, meet one another with courtesy, support one another in trouble. Some three thousand characters, the earliest written village covenant in China; 钱穆 called it a spiritual constitution; the form reached Vietnam and Korea by the fifteenth century. 朱熹 abridged it as 《增损吕氏乡约》, and it entered 宗族规约 (zōngzú guīyuē), lineage covenant, practice downstream. The hedge matters and belongs in the same breath: it was later cited and adapted, but a direct chain to specific Lingnan or Wuyi hall rules is not documented.

The same household produced 吕大临’s 《考古图》, around 1092, in ten chapters — the first systematic illustrated bronze catalogue: 摹绘, 释读, 尺寸, 藏家, drawings, readings, measurements, collector notes. A 吕 invented the genre of the catalogue.

The irony: a surname that left no villages in 四邑 left the rulebook those villages’ halls were run by.

The Hinge at 885, and the Trap at 111 BCE

Genealogical tradition says 吕竞茂 (Lǚ Jìngmào), taboo name 占, entered Fujian with the Three Wang in 光启元年 (885), was granted a residence by 王审知, and later moved to 晋江 安海. Every clause sits under “genealogical tradition says” — 谱牒传说 (pǔdié chuánshuō).

Three cracks show. Its earliest surviving text is a mid-Ming Quanzhou 吕 genealogy; no Song or Yuan document supports it; and the arithmetic fails, since 885 to 吕惠卿’s birth in 1032 allows four or five generations, so any genealogy spanning that interval must compress or pad it. “Entering Fujian from 固始 (Gùshǐ)” is a formula dozens of Minnan surnames share, so a genealogy alone cannot separate a real move from Ming–Qing 攀附 (pānfù), pedigree attachment.

What sits beside the tradition is solid. 吕惠卿 (1032–1111) passed the 1057 examination, became 参知政事 (cānzhī zhèngshì) in 1074, and was the second man of Wang Anshi’s reforms before falling out with him; 《宋史》 placed him among the treacherous officials, which is a fact about the record. 吕夏卿 (1015–1068) passed in 1042, was brought into the New Tang History by Ouyang Xiu and Song Qi, and built its tables of imperial and ministerial descent. Both are documented by government gazetteer pages, not by genealogies: the hinge is a tradition, and its descendants are officials in the record.

The trap is 吕嘉 (Lǚ Jiā), chancellor of 南越 (Nányuè), a 番禺 (Pānyú) Yue man who served four kings, killed King Xing and the Han queen 樛氏 in 112 BCE over their move toward direct Han rule, and fell when the Han army destroyed the kingdom in 111 BCE. An indigenous elite adopting a Han surname is evidence against a migration pedigree, not for one, which makes “吕嘉, our ancestor who entered Guangdong” a category error rather than a lineage claim. 吕 was in Guangdong a thousand years before 珠玑巷, and that fact belongs to a different family.

The Road by Sea, and the Charter That Excludes It

A Taiwan volume of 1976, the 河东堂 《吕姓大宗谱》, gives exactly three Guangdong branches — 潮州, 丰顺, 海丰 — a coastal set with no Pearl River Delta branch. That is its structure as catalogued; the 743-page book itself was not consulted.

At 潮州府西街 stood the 敦睦堂 吕氏大宗祠, built around 1742–1744. Temple and genealogy sources record it; details vary and none is independently verified. Its 谱序 records that genealogies had been lost in war and the hall was built to reunite the lineage, and the branches unified their 字派 (zìpài), their generation sequence.

One hall and one sequence held a Teochew-speaking lineage and a Hakka branch together, which indicates that surname affiliation could be organised across dialect groups: the binding force was the surname, not the speech.

潮阳胪岗 (Cháoyáng Lúgǎng) in 潮南区 is the 粤东大宗 (Yuèdōng dàzōng), eastern Guangdong’s great lineage: 133,073 people on a 2003-era figure from local sources, its 吕厝 (Lǚ Cuò) ward holding over six thousand people and a 祖祠.

Three incompatible Song ancestors, all printed, none resolved. The Chaozhou genealogy’s 吕大防 line runs 通 → 英 → 大防 → 景山 → … → 机学, who fled to 胪岗 at the Song–Yuan turn. The union catalogue’s entry for the 1744 hall genealogy says instead 东莱, a seventh-generation descendant of 吕蒙正, moving to Zhangzhou in the Song — a bibliographic attribution, not descent. The 1976 volume routes 吕祖谦 through 吕蒙亨, not 吕蒙正, an inconsistency inside one book’s front matter.

丰顺 sits in the 梅州 Hakka belt, whose wider geography is mapped separately; 海丰 straddles the Hoklo–Hakka line, and its 鹿境 poem reads 「源发蒲田远，流入海丰长」 — 莆田, not 珠玑巷 or 固始, and one branch’s memory rather than a universal origin.

吕 is absent from 罗贵’s 1131 roster of thirty-three surnames and ninety-seven households — and 卢 and 陆 are on it. The two surnames this one is most often confused with are inside the charter, and 吕 is outside. Three failures stand on their own: a Song document calling 南雄 a 府 is anachronistic, since it became 南雄府 only in 洪武元年; the flight is dated 绍兴元年 (1131) in one strand and 咸淳 (1265–74) in the official heritage account, 140 years apart; and the pedigree’s real function dates from 嘉靖十五年 (1536), when 夏言’s memorial, the 推恩令, let commoners build halls and sacrifice to founding ancestors, turning a Song ancestor in 珠玑巷 into a household-registration credential against the boat people and the Yao. The defensible denominator is 103 surnames and 197 lineages.

The doctrine is real: 《广东新语》 says the old families of Guangdong all came from the lane, and the provincial government lists the legend as intangible heritage. This article’s claim is narrow: 吕 is absent from the roster, so inclusion cannot be assumed without further evidence — not that no 吕 family tells the story. A route by sea is what the sources show; no inland lane is documented in them, which does not categorically exclude inland movement. The Fujian–Chaozhou frame carries the corridor’s mechanics.

Where the Surname Is Not

The negative has denominators. 《台山县志》 gives surnames about seven pages, 86 to 92, with no per-surname population table; its village geography ranks 黄 291, 陈 232, 李 222, 伍 170 — a landscape counted surname by surname — and 吕 is recorded in zero modern villages. Because the gazetteer has no such column, no percentage of Taishan can be quoted for 吕, ever. 《开平县志》 (2002), a tally of 292 surnames, names no 吕 聚居村 (jùjū cūn, settlement village) at all, and no Kaiping percentage may be computed either.

The county’s surname column publishes village-level tables; no 吕 page appeared in it as retrieved on 29 October 2026, the re-check being blocked. As retrieved, not as exhaustive. One circulating ranking, from a single unverified web dataset, places 吕 83rd of 156 surnames, with 卢 65th, 陆 68th and 雷 20th. It is not a population number or a percentage, nothing may be computed from it, and its only use here is rhetorical.

The 1963 《台山县各村氏姓索引》 — Index of Clan Names by Villages for Toishan District, compiled by the Control Unit of a US consulate general in Hong Kong, published by the Association of Research Libraries in 1973 — carries exactly one village with the surname 吕 in the whole county: 岭安 (Lǐng’ān), filed in 第3区 西南江乡 under the sub-heung 横江乡, its compilers’ working note marking the entry 待证. The filing is confirmed at transcription level, by the Friends of Roots VillageDB, a complete data entry of the same volume, which returns the same single row. The page image was not obtained — HathiTrust is search-only, FamilySearch gated — so nothing here is a page read, and 待证 is attributed to the compilers, not observed.

What the line supports is a 吕 侨户 (qiáohù, an emigrant household) in Ling’an. What it does not support is that Ling’an is a 吕 village. Ling’an is not mapped to a modern village and is not an ancestral village.

Its frame has dissolved. The district system died with the 1958 公社 (gōngshè, communes), and half of 三埠 (Sānbù) was legally Taishan until the town was created in November 1949 — so a 吕 household there was 台山籍 and its emigrants appear in American records as Taishanese, like the county’s other emigrants. The filing unit 横江乡 is identified, on one web source, with today’s 斗山镇 横江村委会: a claim about a district, not about Ling’an.

The absence is a finding, not a gap. If a 吕 village were found in Taishan or Kaiping, none of the 通谱’s three branches would explain it; it would mark an independent root. Every negative here is phrased as the corpus allows: the gazetteers consulted contain no 吕 village.

The Name the Clerk Heard

The reason a 吕 family is hard to find is linguistic before it is archival. In 四邑 pronunciation 吕 is Leoi⁵, written Lui; 雷 is also Lui; and 卢 is close enough that the three were interchanged in genealogies, on spirit tablets and on remittance envelopes. The dialect’s five tones — 33, 55, 11, 21, 32 — with frequent 变调 (biàndiào, tone sandhi) let one family’s successive manifests read Lui, Loy, Leu, Looi and Lu.

The archive concedes the collision. A FamilySearch-catalogued 吕 genealogy is titled in English “Genealogy of Lui family of Weijie gong branch” — the cataloguer chose the letter, not the family.

So the rule: a manifest line reading Lui cannot be assigned to 吕 without the birth village and the shipboard kin group. Under the exclusion era of 1882 to 1943, paper-son practice — 冒籍 (mào jí), 纸儿子 (zhǐ érzi) — could obscure surname recovery further: a 吕 who entered as a 陈 left an entirely 陈 archive, as the paper-son material and the records guide set out.

The first tool is the generation poem. As the 1976 volume’s catalogue transcription gives it, the 竞茂公派 昭穆 (zhāomù) opens 茂启余图，蒙简天心，卿清溪语，开基发祖 … — the ellipsis is the transcription’s. One character places a sender in a branch and a generation at once, so a cousin in Penang can rank himself without the book, which is how a 谱 is read. Branch sequences diverge — 胪岗’s 永思堂, the 海门 line — so a recipient’s name often fails to match the genealogy name and must be resolved through a 号, a 乳名 or a 排行. Genealogies record women only as 吕门某氏, so a letter to a wife has no matching entry at all.

The second is the 侨刊 (qiáokān), the emigrant periodical: Taishan’s include 《新宁杂志》, founded at 台城 on 5 February 1909 and described by government sources as China’s first. Three columns matter. The 捐资芳名录 prints name, village and amount, where a Ling’an donor would appear as one line of figures. The 村讯 lists emigrants by township and village, a continuous per-village register. The 寻人启事 is a transnational missing-persons column. They confirm existence, not population.

No 吕 remittance letter — 侨批 (qiáopī) or 银信 (yínxìn) — has been located. The genre is UNESCO-listed, but survival tracks surname size: thick for 陈, 黄 and 李, thin for a small surname. The envelope is what the record should contain; none is catalogued. The genre itself is documented elsewhere.

A Surname Too Thin to Build a Hall

In the Wuyi sources consulted, 吕 emigrants joined 地缘会馆 (dìyuán huìguǎn), territorial associations — 冈州, 肇庆, 宁阳, 四邑 — rather than a 吕 surname association, because a surname numbering a few thousand cannot fund a hall. No Wuyi 吕 surname hall is documented in the sources consulted, and this is not a surname-wide finding. The clan associations that did form show what such an institution needs.

The counter-case: the Singapore Lü Clan Association, registered in 1950, counted some 220 to 250 member families after sixty years. On 2 January 1993 it hosted the first Asian Lü clan gathering, its chairman 吕良山 noting that the 吕 clan, though not a great surname, had not gone quiet. Its own history page supplies the explanation: 吕氏不是大姓，在新加坡，吕姓家庭并不多 — the 吕 are not a great surname, and in Singapore there are not many 吕 households. A Thai 吕 association is reported too; details are not independently verified.

Inference, labelled as one: subscription scale may explain why a small Wuyi surname could not sustain a hall. Where 吕 concentrated — Singapore, Thailand, the Teochew coast — 吕 built associations; in 四邑 they joined someone else’s.

The Wuyi workaround was 联宗 (liánzōng), an alliance of surnames: 雷, 方 and 邝 shared 溯源堂 (Sùyuán táng) and its 《溯源月刊》, founded at 台城 in 1926, to amortise one publication across three surnames.

No 吕 genealogy is catalogued for Taishan — none catalogued is not none existing. Guangdong 吕 genealogies do survive: the Chaozhou hall genealogy of 1744, and a 1977 printed genealogy of the Longpan 吕 of Huiyang.

The 1976 河东堂 volume was published in Taiwan, edited by 卢俊华, a compiler surnamed 卢, with 河东堂 used as a publisher’s name. Its catalogue record gives 4,070 years from the Yu–Xia transition to 1976 and sixty to seventy generations carried. At thirty years a generation that span implies about 136 generations — this article’s arithmetic, not the book’s — so sixty to seventy generations works out at 58 to 68 years apiece. Its upper section is necessarily padded, and the 接谱 (jiēpǔ) method that stitched the Chaozhou, Fengshun, Haifeng and Taiwan branches onto one trunk adjusted counts at each joint. The volume itself was not consulted; structure, poem and counts come from catalogue transcriptions.

So a genealogy is not discovered history; it is edited history. That is a statement about the joints, not the branches. Chaozhou, Fengshun and Haifeng are not thereby false; the seams are, and the book is a print commodity of the roots boom, not a scandal.

Spouse, Née X — and What to Look For Instead

The genealogy enters a wife as 配某氏 (pèi mǒu shì), “spouse, née X”, with no name, no birth, no death and no natal village. The remittance envelope names her 收件人 (shōujiànrén), recipient; the periodical’s obituary names her 丧主 (sāngzhǔ), chief mourner. Three documents, three roles, one woman never named. At the wedding, if the groom was already abroad, a rooster stood in for him at the bow — 隔洋婚 (géyáng hūn), marriage across the ocean — and the years after are what the periodicals called living widowhood, 守活寡 (shǒu huóguǎ). The archive yields no number for those years; it yields the word.

The irony returns here. A surname that left no villages in 四邑 wrote the covenant those villages’ halls were run on — an influential later model, as hedged above, not a documented chain. The Hedong plaque names a Ji-clan fief in Shanxi; the 1976 generation poem ranks a cousin in Southeast Asia; the donor roll prints a Ling’an donor as one line of figures. None of these objects is a village.

What a reader has is a search order, with the surname archive as the map. First the generation character, which ranks a sender without the book. Second the 侨刊 columns, a continuous per-village register. Third the 政区沿革 (zhèngqū yángé), because a village name in a 1963 index is not a place on a 2026 map — the practitioner’s method is the same one. Fourth the destination end, by village and kin group and never by surname alone.

The present tense: 胪岗’s 祖祠 still stands, and the 1742 hall’s generation sequence still ranks a Chaozhou 吕. The 1963 index is 218 pages a reader can open tonight, free, with a FamilySearch account. Consumer DNA testing is corroboration, never proof. This is a place to look and a letter to check, not an itinerary.

Two conditional sentences close it. If the family says 珠玑巷, it is repeating a charter 吕 is not in — and the doctrine being repeated is real, which is why the correction has to be gentle. If the family says 台山 and the record says nothing, the family is not missing paper. It is filed under a letter that also belongs to 雷.

Continue Reading

• The Lu (陆) Surname: From Yamen’s Loyal Minister to the Wuyi Diaspora — the character most often mistaken for this one, and the article that owns its traffic.
• The Lei (雷) Surname: From Ancient Fangshan to the Wuyi Diaspora — the other Lui, and the reason a manifest line cannot be read alone.
• The Jiang (姜) Surname: From the Yan Emperor and Jiang Taigong to the Wuyi Diaspora — the clan name behind 姜姓吕氏, and the grammar this article borrows rather than rebuilds.
• Mao Surname (毛): The Hundred Thousand Who Were Never There — the same discipline applied to a different absence: registers, arithmetic and what the corpus will not say.

Stay Connected

Follow RootsOfChina on Facebook and Instagram for new article updates, behind-the-scenes from Kaiping, and stories from the Chinese diaspora.

Sources

Canonical texts

• 《史记·齐太公世家》 — 「虞夏之际封于吕」; 「本姓姜氏，从其封姓，故曰吕尚」. Chinese Text Project.
• 《尚书·吕刑》 — 「五刑之疑有赦，其审克之」; also transmitted as 《甫刑》. Chinese Text Project, Marquis of Lü on Punishments.
• 《诗经·崧高》 — 「维申及甫，维周之翰」.
• 《左传》 — 僖公十五年 吕甥; 吕相 绝秦 (578 BCE); the Chu campaigns against 申 and 吕.
• 《魏书·官氏志》 — 叱吕氏 changed to 吕.
• 《宋史》 — 吕蒙正, 吕夷简, 吕公著, and its own ambiguity on the family link.
• 朱熹《增损吕氏乡约》; 吕大临《考古图》.

Gazetteers, official records and catalogues

• 《读史方舆纪要》 卷四十一, 霍州吕城条.
• 《水经注》 淯水篇; 《元统志》; 《大清一统志》 — the 董吕村 triangulation.
• 《元和姓纂》 — the Tang codification of 郡望.
• 《台山县志》, 台山县方志编纂委员会, 广东人民出版社, 1998 — surname pages 86–92; village ranks; no per-surname population table.
• 《开平县志》, 开平市地方志编纂委员会, 中华书局, 2002 — 292 surnames; no 吕 settlement village.
• 《台山县各村氏姓索引》 / Index of Clan Names by Villages for Toishan District, Control Unit, Consular Section, American Consulate General, Hong Kong; Center for Chinese Research Materials, Association of Research Libraries, 1973, 218 pp. — FamilySearch item 489257; HathiTrust record 002245737. Friends of Roots VillageDB is a complete data entry of the same volume.
• 广东省人民政府, 珠玑巷人南迁传说 — the intangible-heritage listing for the lane’s migration legend.
• 中国人民大学清史研究所, 《道光〈新会县志·图说〉所载姓氏分布特点》 — the 33-surname roster, 卢 and 陆 present, 吕 absent.
• 晋江市人民政府, 吕夏卿 — the 1042 进士 and his work on the New Tang History’s genealogical tables.
• 固始县人民政府, 三王入闽 — the shared-migration frame behind the 885 tradition.
• 广州市人民政府, 夏言与推恩令 — the 1536 memorial that let commoners build halls and sacrifice to founding ancestors.
• 中新网, 26 June 2014 — the two candidate 吕 state sites; 「尚待考证」.
• 台山市人民政府, 台山姓氏 column — village-level surname tables; no 吕 page as retrieved.
• 新浪转大河报, 2006-01-24 — 董吕村 to 董营村, 王村乡.

Studies and reporting

• 吕耕, 《吕姓溯源》, 文物出版社, 2017.
• 徐玉清, 《百家姓书库·吕》, 陕西人民出版社, 2002.
• 骆伟, 《岭南姓氏族谱辑录》, 广东人民出版社, 2002.
• 刘进, 《银信与五邑侨乡社会》, 广东人民出版社, 2011.
• 梅伟强、张国雄, 《五邑华侨华人史》, 广东高等教育出版社, 2001.
• 梅伟强等, 《台山历史文化集》 第六编《侨刊乡讯》, 中国华侨出版社, 2007.
• 《中国家谱总目》 吕姓 entries — 潮州 大宗祠族谱 (1744); 惠阳 龙蟠吕氏家谱.
• 《吕姓大宗谱》, 卢俊华 ed., 河东堂, Taiwan, 1976, 1 vol., 743 pp. — quoted from catalogue transcriptions only.
• 新加坡吕氏公会, 本会简史 — registration in 1950, membership, and 「吕氏不是大姓」.
• 《联合晚报》, 2 January 1993 — the first Asian Lü clan gathering, 会长 吕良山 (NLB digitised newspapers).
• 中新网转《联合早报》, 2010-12-06 — the Singapore association’s renovation and membership.