The Family Tree Problem Solver: Tried-And-True Tactics for Tracing Elusive Ancestors (9 page)

5. MIGRATION CAN OCCUR IN THE ELDER YEARS.
If you are having difficulty establishing the death date of an older person who lived in one community for several years and then disappeared, most likely he or she migrated and then died in the home of a child — usually a daughter.

6. A CHANGE OF NAME MAY INDICATE A DEATH.
If “Peggy,” who was named after her grandmother, suddenly becomes known as “Margaret,” this may indicate the death of the older woman for whom the child was named. “Polly” may become Mary, “Betty” may become Elizabeth, and so on. Watch for the changing of a title from junior to senior. As you may know, before the twentieth century,
junior
did not necessarily indicate that a man was the son of the man with the same name who used senior. It demonstrated only that there was an older man of the same name living in the community. See chapter eight, “Sorting Individuals of the Same Name,” for a more extensive discussion.

7.
Watch for legal transactions of the off spring. The division of lands previously held is an indication of death, and if the lands are sold and the heirs leave the community, both parents are likely to be deceased. Land sold “in the interest of Samuel Connelly, a minor” may be a clue to the death of a parent. Note the absence of one-third of the land of the deceased; it may be held for the widow during her lifetime, and then dispersed at her death. When the land is divided into parts and all the children are accounted for, the wife of the deceased man has predeceased him.

1.
Probate records should be checked carefully. Note wills in which the husband (or anyone with his surname) is named as a “legatee,” and look for probate in the county of the husband's or wife's residence at the approximate time of marriage.

It was common during the eighteenth century for the witnesses of a will to be one person from the wife's side of the family and one from the husband's. This is not a hard-and-fast rule, but one certainly worth investigating. Often guardians appointed for the children will be a member of her family as well as one of his.
Guardians are almost always family members.
If you had dependent children, wouldn't you most likely choose a family member to look after any estate your children might inherit? The only exception I have noticed to this rule also gives a clue to marriage: the widow's new husband, the children's stepfather, becomes the guardian.

2.
Look for deeds in which the husband was the grantee of a land parcel of unknown origin. Perhaps his wife brought the land to the marriage. Before you jump to this conclusion, however, be sure to check the many ways people may acquire land other than in fee simple. Check delinquent tax sales, civil suits, entitlement (such as military bounty land), federal land sales, and other nontraditional methods for acquiring land that have nothing to do with inheritance. Look for situations in which the husband was a co-owner with others who were not his brothers and sisters or their spouses. This may indicate lands held in common with his wife's siblings.

3.
Quitclaims executed by a husband who may be acting in his wife's place.

4.
Watch for the sudden prosperity of a male ancestor that you can't account for. This may indicate inheritance from the bride's side of the family.

5.
Any deeds involving a woman are worth investigating. She may be selling property she inherited from her father or husband; quitclaiming for the same reason; selling land with a son or brother; or selling lands to a child.

6.
Unusual names that may be surnames can often be clues worth searching. Pressly Glenn, Holloway Cobb, Avery Morgan, and Bailey Bond were all named for the mother's side of the family.

7.
Marriage contracts, often found in deed books, trust deeds, intestate land divisions, deeds of gift, equity and probate court records, quitclaims, and property settlements in divorce cases are all possible sources for marriage records and identification of a spouse. The following deed of gift does not name the husband, although most gifts to married women do. This one is especially nice, as it also names three grandchildren. As the deed was recorded in Missouri, we can assume the daughter was living there.

The following trust deed provides the residence of the father and the Tennessee residence of his daughter, her husband, and her later residence in Missouri. It can be assumed that Isaac Fulkerson was then in Virginia with his probable grandfather, Reuben Bradley, but that he either lived in Missouri or was planning to move there. Reuben's daughter, Sarah Fulkerson, and her husband Frederick were planning to move to Missouri, probably to join their son and perhaps other relatives.

8.
Cemeteries can provide inferences of marriage from inscriptions, family groupings, and even types of tombstones. A young woman (age twenty-two to twenty-eight) in a family burial plot who has a different surname from the others buried there was probably a daughter who married, died young (perhaps in childbirth), and whose husband then remarried. Look for adjoining stones with different names, but of the same shape, type of stone, and style of carving (see
Figure 2-14
).

These tombstones were probably purchased, carved, and set at the same time, most likely when Lonnie died. We can assume that Virginia was first married to William F. Huffman, but by 1897 she was married to H.S. Miller.

A “stranger” in the family plot may actually be a relative who can provide clues to a woman's maiden name. Two children lie in the Strain family plot in a Monmouth, Warren County, Illinois, cemetery. Nathan Brown, a man of eighty years, was also buried in that plot. He died over twenty years after the Strain family had left Warren County. I had assumed James Strain had sold part of the plot to a stranger who needed a burial place. Wrong! Nathan Brown was Nancy Strain's younger brother.

Sometimes no stones exist, but there may be a record of purchases of plots. Knowing that an ancestor purchased a plot where someone whose name you do not recognize is buried may indicate the burial of an in-law or other relative. I found the unmarked graves of my great-grandparents in Prosser, Washington, because their infant grandson, who would have been my great-uncle, was in the same plot. His gravestone led me to the records that indicated who had purchased the cemetery plots.

Figure 2-14
William F. son of G.W. & Virginia Huffman b. Dec. 25 1887 d. Mar. 17 1888. Lonnie son of H.S. & Virginia Miller b. Jan. 25 1897 d. Feb. 6 1897.

Clues to the
date of a marriage
are:

a.
the birth of the first child.

b.
the sale of a parcel of land from a father to his son or son-in-law, especially if the sale was for a small amount of money, such as $50 (know the going rate for land in the area at the time to verify this).

c.
the acquisition of a small parcel of land by a young man — he has either just married or is likely to be intending to.

d.
movement on the Pennsylvania tax list from freeman (or inmate) to regular poll tax for a white male

9.
Correlating two dates, one in which the woman was single and one closely following in which she was married. As an example, I learned from a deed dated 14 November 1700 that Hannah, the wife of William Sumner of Middletown, Connecticut, was the daughter of Daniel Henchman of Boston. But by 1700, Hannah had been married a long time — her first child was born 22 November 1679. Could I come closer to estimating the time of marriage? Deeds again provided the answer. On 15 August 1678, along with her brother Hezekiah Henchman, Hannah witnessed a deed as Hannah Henchman, and on 4 April 1679, she was a witness to her grandmother Elizabeth Clement's deed of gift as Hannah Sumner. So Hannah Henchman was married to William Sumner between August 1678 and April 1679. In fact, unless Hannah and William jumped the gun by necessity, they were married between August 1678 and February 1679, as her first son was born in November.

Locating births, deaths, and marriages when no vital records are available requires studying a multitude of records. It demands the correlation of data rather than finding that one record of absolute proof. It means the researcher makes a logical and chronological study of the records to bring together the myriad events that re-create an individual's life — one begun by birth, connected by marriage, and ended by death.

¹
Donald Lines Jacobus, “Retroactive Dates and Places,”
The American Genealogist
34 (January 1967): 31–35.

²
M. Halsey Thomas, ed.,
The Diary of Samuel Sewall 1674–1729
2 vols., (New York: Farrar, Straus and Giroux, 1973), I:80.

³
Francis G. Walett,
The Diary of Ebenezer Parkman 1703–1782
(Worcester: American Antiquarian Society, 1974), 128.

⁴
“Some Obituaries and Death Notices,” abstracted from the
St. Louis Christian Advocate
of the Methodist Episcopal Church, South, various volumes of
Missouri Pioneers
compiled and published by Nadine Hodges and Mrs. Howard W. Woodruff.

⁵
“Buncombe County, North Carolina, Special Court Proceedings, 1798–1812,” 326. FHL film 410980.

A
ncestors sometimes do not appear where we expect them to in the federal census. Too often a genealogist assumes that the enumerator “missed” the individual or family sought and abandons the search. Let's explore some of the common reasons you may not be able to find your ancestors in census reports, and I'll suggest methods to overcome this difficulty.

The federal census is a resource genealogists depend upon. It is easily accessible through microfilm, interlibrary loan, and online through such websites as HeritageQuest and
Ancestry.com
. Because it is theoretically a count of all the people then living in the United States, we deservedly have high expectations of finding the person we seek. We also know that the census contains information valuable to the genealogist. It is more than just a list of names; it contains data needed for establishing relationships, and clues to further research.

In fact, when one genealogist asks another for help, the first question the helper usually asks is “Did you find those people in the census?” A census, especially those from 1850 on, puts a person in time and space, gives him an approximate age, places him inside a residence (usually with other people who are significant to him), notes his place of birth, and offers other valuable information used by astute researchers. But what happens when the individual is not found in the expected census? The most common step that genealogists take is to move back another decade and try to place the person ten years earlier. Often this fails. Why can't the individual be located in the census where he should be? The easiest answer is that the census taker missed him. That reply — made all too often — is frequently false. Too seldom does the genealogist ask
how else
the individual might have been listed,
where else
he might have been listed, or — if he was indeed missed by the census taker —
why
he was missed, and who else might have been missed as well.

Before I develop some of these issues and list some of the reasons that individuals do not appear in an expected place in a census, first
let's transform ourselves from the person searching into the person sought.
Think about where you probably appeared on the censuses taken during your lifetime. This may give you some insight into why your ancestor is difficult to locate. Track your own life and reflect where you were or
should have
been during each decade the census was taken. Do you remember you or your parents filling out the census form? Would you be easy or difficult for a future researcher to find? Why?

The census is usually taken by a certain day — even though the actual day the census taker visits and the date the individual fills out the form may vary significantly. Those dates, however, usually fall in the summer. I have tracked my own life showing where I should appear on the various decennial censuses.

Two significant items emerge from the above for the purposes of our discussion. First, individuals tend to be the most mobile between the ages of eighteen and forty. Of course, the childhood years are when the genealogist is hoping to find and connect a person with his or her parents. For two of the above censuses I should have been counted with my parents, but for the 1960 census, we were in transit. After finishing my education, beginning my career, and marrying, my life became more stable — as it does for many people — so I remained in one location for over thirty years.

Often genealogists begin to reconstruct a person's life from the information we find on the census. Nevertheless, the census provides just a skeleton of a family. The census is only one step in the research process. Often you find the information you seek only after have you have used the information on the census to lead you to other records about the family.

Individuals are particularly hard to find before 1850, when the census listed only the heads of households. Using information I gained from a study of one thousand early Southwest Missouri pioneers, I believe I have found 141 of those pioneers who were not heads of households in 1830, but were listed in another household as just a slash mark in the appropriate age category. I've determined that 108 were living with their fathers, twenty-two with their widowed mothers, five with their brothers, two with their fathers-in-law, three with their stepfathers, and one with his guardian. It was extremely difficult for me to locate those 141 people, so I am not surprised that there are still 271 individuals out of the one thousand whom I can't find listed in 1830. This does not mean that these people were not included in the census. They may be in similar situations to the ones I will describe in the next pages.

In addition to the problem presented when an ancestor is not the head of household in an expected area, an ancestor's mobility may be another major issue for those searching censuses before 1850. We think of our society as a mobile one, but many of our ancestors, particularly those living in the nineteenth century, were just as prone to move as we are. After the War of 1812, the largest migration in the history of America began — just at the time when the census provides only limited information, making it very hard to locate a pioneering ancestor. Of the 1,000 individuals who purchased land in Southwest Missouri between 1835 and 1839, only about 91 percent can be found on the next available census: the 1840. By 1850, more have disappeared from the rolls. Of those who purchased land in Southwest Missouri before 1839, 125 cannot be found anywhere on the 1850 census.

Considering the handicaps of terrain, the lack of decent roads, poor transportation, the isolation of the people, and the rigors of the job, it is amazing that the census enumerators did as well as they did. Today we often hear about undercounting in the census, and no doubt omissions occurred in the nineteenth and early twentieth centuries as well. However, it is important to remember that census takers were paid by the number of people they recorded in a county. For instance, in 1810 the pay for federal census takers was two and one-half cents per person listed. Thus, those taking the census were financially motivated to locate as many people as possible. Too often the genealogist is quick to blame the enumerator.

In fact,
there are three presumptions we frequently make when we cannot find the individual or family we seek.

1. PRESUMPTION OF ERROR:
We presume the census taker was at fault. You will often hear, “A lazy census taker evaluated the neighborhood,” or “The census taker only enumerated some families in the area and ignored mine,” or “That census enumerator was careless and sloppy.”

2. PRESUMPTION OF BLAME:
We presume the ancestor was at fault. “Our family never wanted anything to do with government people,” or “Hiram Hideaway must have been avoiding the census taker,” or “I bet old Hiram told a wild tale to that census taker and didn't answer the questions properly. He didn't think his name or his family was any business of the government.” That may have been true, but old Hiram was taking a risk. It was illegal to knowingly give false information to a census taker.

3. PRESUMPTION OF LOSS:
We presume that no important information can be gleaned from that census if the ancestor we seek is not listed.

If the researcher presumes any, or all, of the above incorrect suppositions, the result is a missed opportunity.

Discussed below are fourteen types of problems that researchers sometimes encounter when using a census. People may appear to be missing because the researcher is using finding aids rather than the census itself, or because they may be “hidden” in an unexpected section. Some people may actually be missing because they were not counted. Examples of these problems and ideas for solving them follow.

1.
What do you do when the ancestor you're researching is not found in a published computerized census index, on a CD-ROM, or with an Internet search? Far too many genealogists presume that when their ancestor is not in the census index, he was not in the census. The census index is a
finding aid
— that is all.
Aid
means
help
or
assistance
, not
replacement
.

Some problems are inherent when using a computerized index. See Richard Saldana's
A Practical Guide to the “Misteaks” Made in Census Indexes
(Salt Lake City: R.H. Saldana & Company, 1987) for suggestions as to how to work with these difficulties. Problems include spelling variations, complete omissions, typographical errors, and reversed first and last names.

If you are working in a small or rural county, it may be easier to just check the census itself than to figure out the index.
John Biggs and William Montgomery were absent from the census index, but easily located on the census in the county expected. Caleb Headlee was indexed as “Hadley”; Jane Cawlfield as “Coffield.” When the actual census was checked and the names placed in the context of the neighborhood, they were located without difficulty. The spellings differed enough to place them in unexpected areas of a large census index, but not so different that they were not immediately recognized when the names appeared in the census itself.

What do you expect to gain when you pick up a census index? Do you expect to find your ancestor? Narrow those expectations. Use the index to help find a
page
, not a
person
. If you find the person and page you need, fine. If you do not locate that individual, you still have no idea whether she was on the census itself. Almost everyone who has been in this field long enough can give an example of finding an individual on the census who did not appear in the index.

2.
New problems arise with using CD-ROMs and Internet databases to search census indexes. When reading text in a census index,
Sameul
and
Goerge
are easily recognized as typographical errors. When you search a computer for a name, however, the computer will not “hit” on a misspelled name if the researcher has typed in the correct spelling. Most census indexes on CD or the Internet do not permit “wild card” searches, which can cause you to miss the individual.

3.
Some compiled census materials are done very well; others are incredibly sloppy. If you are using a published census odds are that, even if you find the individual you are seeking, not all the information included in the census will be included in the publication. Again, if you don't find the individual you are trying to find in the publication, this doesn't mean he wasn't in the county. If the compiler is not familiar with the handwriting of the period, the enumerator's style, or the families in the community, the name may be garbled. Some names and letters are easily misread.

4.
You may think that unusual names would be easy to trace — but not in the census! While common names often can be deciphered even when poorly written, unusual ones can't. Many of these names could be determined by a knowledgeable researcher who is familiar with the family and the community, but they are not likely to be found simply with a computerized search.

Some initials and letters are harder to distinguish than others. The names
Daniel
and
David
may be difficult to differentiate; the letters
M, H, N
, and
W
may also be difficult to discern from one another.