The Problem of William H. Smith (& Family)
When he enlisted in the 36th Iowa Volunteer Infantry on 11 August 1862, William H. Smith, a 39-year-old married resident of Monroe County, Iowa, gave his date and place of birth as 9 July 1823 in Boone County, Illinois. Boone County, located near the center of the Illinois-Michigan line, was not created until 1837 -- 14 years after William's birth -- but we may assume he was simply identifying his birthplace by the name by which it was known in 1862. It seems very unlikely, however, that a pregnant woman would be out in what was then still unsettled wilderness. (The Winnebago Indians of northern Illinois were enthusiastic participants in the Black Hawk War of 1832.) And, according to a history of Boone County published in 1881, the first known white settlers didn't arrive in the neighborhood of Belvedere (later the seat of Boone County) until the summer of 1835. Still, William was explicit more than once in naming his birthplace. The question is, where was he between his birth and his appearance in Monroe County, Iowa, in 1853?
Question: Is it possible he actually was born in Boone County, Indiana? This would require consistent error or misrepresentation on everyone's part, however. But this Boone County was created in 1830 and was already being settled in the early 1820s. This would place his family only two counties to the east of Fountain County, the import of which will be noted later.
There's also a Boone County, Missouri, and lots of people moved back and forth between Iowa and Missouri. The Missouri county was organized in 1820, with settlement beginning a decade before that, which is good for William. Moreover, a recent discovery of a possibly relevant family in the 1850 census makes this alternative even more likely; more on that below. Still, why was William consistent in saying he was born in Illinois?
The earliest known record for William H. Smith is the license for his marriage, on 30 June 1853, to Caroline Rowles of Troy Township, Monroe County. She was the 21-year-old daughter of David H. Rowles and Rebecca (Clark) Rowles, who were very early settlers and prominent citizens of the county. David Rowles had been a hotel-keeper and justice of the peace in Fountain County, Indiana (where the five youngest of his nine children all were born), before moving to Iowa in 1844. One of Rowles's sons-in-law, John Nebeker Massey (who married Matilda Rowles), was a surveyor who laid out the site of Albia, the seat of Monroe County, and David's sons, Oliver Perry and Anson T. Rowles, held most of the township and county elected offices. Oliver P. Rowles also served in the Iowa state legislature during the Civil War.
William Smith was 30 years old when he married Caroline, which suggests that it may not have been his first marriage; the license does not say, and no evidence has been found one way or the other.
While William was in basic training at Benton Barracks in St. Louis, he wrote home on 28 November 1862 -- not to his wife and not to any blood relative, but to his wife's brother-in-law, John N. Massey, whom he addresses as "Brother John." (Of course, he may have written other letters, but this is the only one that was preserved, by John Massey's descendants.) He describes the camp, which had been set up at the city Fair Grounds, sketches a map of its layout (perhaps of special interest to John Massey, a surveyor), and gives special attention to the sad condition of the surviving ornamental trees, most of which had already been cut down by the troops for firewood. (This seems a reasonable concern for a carpenter, which was his stated occupation at the time of enlistment, as well as in the 1860 census.) He also mentions various mutual acquaintances in his unit and visitors to the camp from back home, but investigation has not shown any of them to be family-connected. Almost as an afterthought, he asks John to give his love to Caroline. While its spelling and capitalization are eccentric, the letter's penmanship suggests at least a modest amount of formal schooling. (This assumes that William did not have the letter written for him by a comrade in camp. . . .)
The 36th Iowa took part in the Yazoo Pass expedition in Mississippi the following February. Much of the regiment spent weeks wading through the chilly swamps, trying (unsuccessfully) to find a back door into Fort Pemberton. William was one of many who was mustered sick during the following month. In April or May, he was shipped back to the army hospital at Keokuk, and died there on 31 August 1863 at the age of 40, of "chronic diarrhoea" (which probably was amoebic dysentery). Because of the length of time he was in hospital in Iowa, it's reasonable to suppose that Caroline was present at his deathbed, though this is not known. He was interred at Oakview Cemetery, on the outskirts of Albia, and a metal G.A.R. marker was later added to his headstone. The following January, Caroline applied for a widow's pension. Her file includes copious documentation of their marriage, their children's names and birth dates (including the original page from the family Bible), and the facts of William's service, but no further information on the other members of the Smith family.
David H. Rowles, age 63, appeared in Troy Township in the Iowa state census of 1856 with an extended household. In addition to his wife, Rebecca, and two teenage daughters still living at home, there were also a number of Smiths under his roof.
William H. Smith was included in the Rowles household, age 32, a carpenter, and 10 years in Iowa; his birthplace was again given as Illinois. Caroline, his wife, was age 24, born in Indiana, and (like her parents and siblings) had been in Iowa for 12 years. Their first two children, Emma and Edward, also were listed, aged 2 years and 1 year.
Also present in the Rowles household (listed next after William, in fact) was Milford C. Smith, age 19, also born in Illinois, and also in Iowa for 10 years. It seems apparent that Milford was some kind of relation to William, though the 13-year difference in their ages may be extreme for siblings -- especially when there were no other obvious siblings present. Nor was he young enough to be William's son by an earlier marriage. Most likely, he was William's nephew, for reasons discussed below. One other person was in the Rowles household in 1856: C. M. Smith, an 11-year-old male born in Iowa. This was almost certainly "Charles M. Smith."
Charles M. Smith, age 18 years, mustered into the 22nd Iowa Infantry at Iowa City on 9 September 1862 -- probably his first opportunity to get in on the War based on his age. On 3 January 1864, having fought at Port Gibson and Champion Hill, and having participated in the sieges of Vicksburg and Jackson, he died of typhoid at Mustang Island, Texas. His effects (including $2.98 in cash) were shipped back to Iowa with his body and he also was buried in Oak View Cemetery, very near William's grave.
Not far away from David Rowles in 1856 lived John and Prudence Lower, age 70 years and 65 years. The Lowers had long been close friends of the Rowleses, both in Indiana and in Iowa, witnessing each other's legal documents, attending the same Bible classes, and signing the same political petitions. The Lowers were even earlier pioneers than the Rowles family, in fact, having arrived in Iowa Territory in 1842. They seem also to have been well acquainted with the Smith clan; their son, Tarkington S. Lower, was married to Mary Margaret Smith (age 17 in 1856), who said she had been born in Iowa. No marriage license or certificate has yet been found in Monroe County and only some of the records of adjacent counties have been available for searching, though the Lowers seem to have spent their whole Iowa careers in Monroe County. (But see below.)
The license for the marriage of Oliver Perry Rowles and Louisa Lower in December 1845 is the second listing in the earliest Monroe County register. Also living in the Lower household in 1856 was Oliver Smith, age 14 years, born in Iowa, and almost certainly Mary Margaret's brother. He will be discussed below in a different context.
The Rowleses and the Lowers lived on neighboring farms only a couple of miles from Albia. In the town itself in 1856, there resided a merchant named H. H. Markham, age 37, born in New York, and his wife Hannah. They seem also to have kept a rooming house, since the census-taker listed 15 persons in the household, all adults, of assorted ages and occupations, and nearly all of different surnames. Among them was Milford Smith, age 20 years, born in Iowa, for whom no occupation was listed. With such an uncommon given name, it seems likely that Milford moved out of the Rowles household between the rural and town enumerations and was double-counted. Either his birthday had just occurred or his age is in error -- as is, probably, the idea that he was born in Iowa in 1836, which would have been extremely early. (I have found in the past that census information collected for rooming houses, hotels, hospitals, and other places where strangers live together temporarily are far more likely to be erroneous than for family dwellings.)
On 13 August 1861, David Rowles wrote his will and made the expected provisions for his wife, his children (all of whom are explicitly named), and his grandchildren (some of whom are identified by name). Then he added the following:
In the event, David Rowles lived another seven years, not dying until 14 July 1868, by which time circumstances had changed and not all the heirs were still living. But who was Emaranda Smith? At this point, the obvious conclusion is that she was the late mother of the six young Smiths named in David Rowles's will as her heirs -- but in that case, who was her husband? Presumably, he also was deceased, but David Rowles did not name him. Nor did he indicate that Emaranda was a Rowles by birth. In any case, all of David's children are known from multiple sources (especially his War of 1812 pension application file) and there are no suspicious gaps in the "ladder" of their birth dates. All his nephews and nieces in 1861 also are believed to be accounted for.
None of these Smiths were living with David Rowles in the 1850 federal census. Only Charles M. Smith, age 15 years and born in Iowa, was living with the Rowles family in 1860. Only in the 1856 state census, as noted above, was there an explicit association, with three of the heirs of Emaranda Smith in the Rowles household, and two others living with John Lower, practically next door.
David M. Smith, named in David Rowles's will (and possibly named for him), has not yet been definitely located in the 1856 or 1860 census; given his position in the listing of heirs in the Rowles will, he may be the eldest of the six and probably was already out on his own in 1856 -- or, he may not be a son at all, but a younger brother of Emaranda's husband (and still one of her heirs).
Milford C. Smith, the oldest of the located heirs, was born in 1837 in Illinois (according to the 1856 census). If the usual conditions prevail, and if Emaranda Smith was indeed his mother, she would have been born before about 1818. If she was also the mother of William H. Smith (even though he was not listed as an heir, and David Rowles explicitly said "being all her heirs"), her birth date would be pushed back by a decade and a half and the question would then arise as to the lack of children born between 1825 and 1835. This would be unusual, even if she had been twice married. This suggests that William and Emaranda were more likely siblings, with only about five years' difference in their ages. If so, then she also should have arrived in Iowa about 1846 -- but, again, no one of her name appears in the 1850 census of Monroe or any adjacent county. No probate record created in Monroe County between 1850 and 1860 includes her name as a principle, heir, or witness (according to the published abstracts), and there is no surviving cemetery listing for her in Monroe County. (All cemeteries in the county were canvassed in the 1930s by the WPA and later by the county genealogical society, and the listings in both cases have been published.) (But see Additional Research, below.)
One other piece of relevant correspondence was preserved in the Massey family: A poorly penned plea for help, dated 26 November 1866 at St. Joseph, Missouri. It was addressed "Dear Uncle" and was signed only "R E Smith." The text is as follows:
The letter was sent to John N. Massey, husband of Matilda Rowles. The sender could only have been Rachel Elvira Webb, daughter of the late Julia (Rowles) Webb, who was Matilda's younger sister. John Massey was indeed Rachel's uncle, by marriage. "Uncle Oliver" was either Oliver Perry Rowles, older brother of Matilda and Julia -- or Oliver B. Smith, who was probably an older brother of Charles M. ("Charlie") Smith, who had recently died in the War. But which Smith did Rachel marry? (See Additional Research, below.)
A search of Monroe and the adjacent counties in the 1860 federal census failed to turn up either Oliver B. or Milford C. Smith. Going farther afield, however, they were finally located . . . in Ottumwa, in the Wapello County Jail. Criminal court records in Wapello have not been found (and probably no longer exist), so it's not known what their offense was, but it's unlikely to have been something as ordinary as drunk-and-disorderly since both young men were imprisoned together. In any case, given the circumstances of Rachel's letter (written six years after they were in jail), one of these two ne'er-do-wells most likely was her husband.
"Charlie" (whose money "Uncle Oliver" had) was almost certainly Charles M. Smith, William H. Smith's presumed younger brother (and therefore Oliver's nephew by marriage -- his sister's husband's brother). The money may refer to an inheritance from Emaranda, or it might be accrued pay from the army; his death file is equivocal on the question of accounts due him. (It could not refer to the bequest in David Rowles's will, since David was still alive in 1866.) (See Additional Research, below.)
But none of this explains the familial responsibility that David Rowles obviously felt for Emaranda's presumed children. If an earlier connection (other than William's marriage to Caroline Rowles) exists between the Rowles and Smith families, it has not yet come to light. One possibility may exist, however. Back in Fountain County, Indiana, previous residence for about 20 years of the Rowleses and the Masseys (and possibly the Lowers), a marriage was recorded on 1 May 1834 between "Haydon Smith" and "Marianda Rowles." Information about this Haydon/Hayden is completely lacking, and the Smith surname may be coincidental, but Marianda must have been related to David H. Rowles, as the latter's family was the only one of that surname in the state of Indiana at the time.
And, in fact, David Rowles had two daughters indicated on pre-1850 censuses in Steuben County, New York, and in Fountain County, Indiana, whose names are unknown. One was born between 1815 and 1818 and appears by age group on the 1830 and 1840 census listings for David Rowles; she would have been 16 to 19 years of age in 1834. She does not appear in the Rowles household in Iowa in 1850. (The other daughter was born about 1820 and is accounted for on the 1830 and 1840 censuses; she would have been only about 14 at the time of the wedding in 1834. Most likely, since there is no unaccounted-for Rowles female in the 1850 census, and no other Rowles female in the Monroe County marriage records for that decade, she simply died between 1840 and 1850.) (See Additional Research, below.)
There is a "Hayden Smith" in Van Buren County, Iowa, in the 1840 federal census with four children under 5 years of age and one child age 5 to 10. There are also two "extra" males in the household, one age 20 to 30 (the same age group as Hayden himself) and one age 15 to 20. The latter is much too old to be his offspring but is just the right age to be William H. Smith. Hayden also appears in Davis County, Iowa (created in 1843 from Van Buren) in the 1847 state census with a household of five persons, but their age and sex are not indicated.
Some confusion is added by the discovery of two more marriages for a "Hayden Smith," both in Davis County. The first was to Ellin Chapman on 28 February 1847, the second to Mary Emaline Box on 25 December 1848. Are all three "Haydens" the same individual? (He ought not to have a child, a "Junior," at that date who would be old enough to marry.) If so, did he and Marianda divorce in 1845 or '46, after their son (presumed to be Charles M.) was born? Marianda's relationship to David Rowles has not been determined, but the speculative possibilities are tantalizing. "Emaranda" and "Marianda" are very similar in sound and are both unusual names, and may have been equally difficult for a clerk or census enumerator to record correctly.
One final direct clue exists for Emaranda Smith: Another marriage was recorded on 18 February 1849 in Monroe County, Iowa, between "Mrs. Emaranda Smith" and David Newell, who apparently was a widower. Their listed ages were 32 and 46, so they would have been born in 1817 and 1803, respectively.
However, a "David Newell" also appears with five other males and one female, living in Jackson County (far to the east, on the Mississippi) at the time of the state census in the summer of 1849; this may or may not be the same person. A search of the 1850 federal census found David (age 48) living in Davis County in the household of John and Lucy Richardson (ages 29 and 35), with -- in addition to two Richardson children -- five offspring surnamed Newell, ages 4 to 14. No wife for David was present.
If "Emaranda" and "Marianda" are in fact the same person (and they probably were), the implication is that David Newell's first wife had died between 1846 and 1848, and he had married again to the divorced or widowed Emaranda Smith in February 1849. And then, Emaranda had also died sometime in the following 16 months. And one explicit record and two reasonable suppositions place Emaranda's date of birth in 1817 or 1818. It also seems reasonable to suspect that William H. Smith and Hayden Smith were brothers and that William became acquainted with Emaranda's sister, Caroline Rowles, through Hayden's first(?) marriage.
I wrote and distributed the above discussion to family members and other interested researchers about a decade ago. I finally located a reference in a commercially published local history, the Memorial Record of Iowa (Chicago: Lewis Publishing Co, 1896), stating explicitly that Mary Margaret Smith, wife of Tarkenton S. Lower, was the daughter of Hayden Smith and Emaranda Rowles.
I now believe David Rowles's first, unnamed, daughter almost certainly to be Emaranda, or perhaps "Emma Randa." Which also means that, since William married Emaranda's sister, Caroline Rowles, he could have been Emaranda's brother but not her son. This also would make William probably the uncle of David, Milford, and Oliver Smith (the heirs and presumed children of Emaranda).
I have also researched further the Hayden Smith who married Ellin Chapman and Mary Box, and I am now convinced that he is a completely separate person from "my" Hayden. His children have been identified, they do not include Oliver, Milford, David, or Charles. And he appears never to have had any connection with the Rowles or Lower families, or even with Monroe County.
Following these discoveries, I unexpectedly made contact online with a descendant of Oliver B. Smith, who told me that Oliver, too, had served in the Civil War, but in a different regiment from William or Charlie. I immediately sent for the complete military service and pension files, . . . and fell into a whole set of new questions and puzzles! A summary of this new information follows:
On 26 March 1862, at age 20, Oliver Smith enlisted in Company F, 17th Iowa Volunteer Infantry at Albia. A month later, apparently while still in training, he came down with measles, for which he was treated for two months at the army hospital in Keokuk (where William Smith would die a couple of months later). Rubella can be very dangerous in adults, especially 150 years ago, and Oliver was intermittently ill throughout the rest of the War, as shown by the frequent notations in the regiment's muster rolls that he was "out sick" and "in hospital." According to his later pension applications, he suffered permanently damaged eyesight and sustained lower back problems. The 17th Iowa served in Tennessee, at the siege of Corinth, at the battle of Iuka, in the Yazoo Pass Expedition (where William became ill), at Port Gibson and Champion Hill, the siege of Vicksburg, at Helena and Chattanooga, and in Sherman's Carolina campaign, ending in the occupation of Raleigh. The regiment then moved to the occupation of Louisville, Kentucky, and was mustered out in July 1865. Because of his recurring illness, Oliver apparently missed nearly all of this, but was nevertheless credited with three years' service and received an honorable discharge.
Then things get complicated.
Oliver must have had an itchy foot after the War because, according to his later pension applications and affidavits, he went to Pettis County, Missouri, in 1865 -- to Arizona in 1877 -- back to Iowa (county not mentioned) in 1878 -- to California in 1880 -- and to Richardson County, Nebraska, in 1882, where he remained until at least 1909. All these peregrinations have made it very difficult to track him in the censuses of 1870 and 1880. He apparently moved to San Diego around 1910 and entered the Naval Hospital there in June 1930. He died June 27 of that year and was cremated and interred at Greenwood Cemetery in San Diego. His California death certificate says his parents were "Hayden Smith" and "Emarandia Rolls," so he is certainly the same Oliver Smith mentioned in the will of David Rowles. (This also provides additional evidence of the marriage of Hayden and Emaranda and of Emaranda's maiden name.)
According to several statements in his pension files, however, Oliver was born 22 July 1847 -- which is five years later than the birth date indicated in the censuses of 1856 and 1860. In fact, an apparent clerical error in one of his later pension applications claimed a birth year of 1846 -- which would have increased his pension eligibility by one year, and the Pension Bureau gave him a hard time over the discrepancy. Oliver had to undergo a good deal of correspondence to mollify the Bureau and get the date "corrected" to 1846. But if he really was born five years earlier, why would he not say so and thereby obtain his pension sooner? I believe the 1842 date must be the correct one, the pension file notwithstanding, because as a young boy he almost certainly did not personally supply the information in the 1856 Iowa State Census -- and I find it hard to believe anyone would claim that a 9-year-old boy was actually 14. Nor is it likely that he would have been in the Wapello County Jail at 13. Not to mention enlisting in the 17th Iowa at the age of 15 (and the enrollment muster definitely says his age was "20 years").
I've been assuming, by the way, that Oliver was named in part for his uncle, Oliver Perry Rowles. But his pension file also gives his middle name as "Bronson" -- and it's worth noting that David Rowles's mother (Oliver's great-grandmother), who died in Steuben County, New York, in 1813, was Christiana Branson.
There are problems with his claimed marriages, too. His 1865 marriage to Rachel Elvira Webb ("R. E. Smith" of the anguished begging letter quoted above) is now well attested. And because of all the documented interrelationships among the Webbs, the Smiths, and the Rowleses, it's hard to believe that Rachel's husband could be another "Oliver B. Smith" with the same family connections in the same community.
But according to the affidavits in his pension file, Oliver was married twice: First, to Lucca (or perhaps "Lucia," but later referred to as "Lucy") Edwards, by whom he had three children, of whom apparently only one survived to adulthood (Luella, born March 1874 and still living in 1915). He doesn't supply the date or place of his first marriage, however, saying only that he doesn't remember and that "everything burned up" -- which, frankly, seems a bit suspicious. He also says there was no church record because they were "married by a justice" -- which means there should be a license on file, somewhere. Lucy must have died after March 1874; just how much later depends on Luella's position among the three children. But, according to Oliver, she died in Brownsville, Henry County, Missouri, which is just to the southwest of Pettis County, where he says he was living after 1865.
Oliver's second wife was Jennie Draper -- also no date or place given, but probably about 1880 and probably in Richardson County, Nebraska. He was her second spouse, too. Her first husband, Watson Sterns, died in Richardson County in 1878, only two years after his marriage to Jennie. She gave Oliver six more children, born between December 1881 and June 1893, all apparently in Richardson County. One of these was Daisy (Smith) Fisher, who was living in San Diego in 1930 and whose address Oliver gave to the Naval Hospital as his own. Jennie seems to have lived until 1927 or 1928 because a Physician's Affidavit in the file, dated January 1929, states that Oliver took care of his wife for several years and that he himself had been under medical care for the previous five months, "following loss of wife."
Nowhere in any of this is there mention of -- or room for -- Rachel Webb. But a subsequent re-examination of the Monroe County, Iowa, Marriage Books turned up a marriage between "Oliver Smith" and Rachel Webb, performed by Rev. Barnes of the Methodist Episcopal Church.
So: Was Oliver actually married three times? He specifically states, though not under oath, that "Lucy Edwards was my first wife." But was he attempting in his later years to erase the memory of an earlier wife to whom he had behaved badly? Remember the letter written by a desperate Rachel to her Uncle Oliver (Rowles) regarding her husband, who was in jail? If this were a soap opera script, I might suggest that Oliver either abandoned Rachel and his infant child (subsequently leaving the state on an extended period of wandering), or that he was ashamed of his incarceration -- which would have been his second time in the pokey, after his 1860 sentence in Ottumwa with Milford. It may also be that Oliver and Rachel were divorced, possibly in Missouri, but no such record has yet surfaced.
Finally, in the Bureau of Pensions Questionaire of March 1915, in which he provides much of the data on his wives and children outlined above, he also states that he was born in 1847 in Bloomfield, Davis County, Iowa. The "other" Hayden Smith, the one whom I had finally convinced myself was not my Hayden Smith, was located in Davis County and was buried in Bloomfield, the county seat. Does this doppelganger thing extend over more than one generation?!
Very recently, my middle son was discharged from the army, got married, and (in the process, I think, of becoming acquainted with his new wife's family) developed an interest in his own genealogy. And almost immediately, he turned up a key piece of information I wish I'd had years ago -- the possible identification of our William H. Smith in the 1850 census, in Clark County, Missouri. Note that Clark County is in the far northeastern corner of Missouri, adjacent to Iowa. Kahoka, the seat of Clark County, is less than 100 miles from Albia.
Since John Cross appears between the two Smith families, is his wife a Smith? The age and birthplace fit -- but two Elizabeths of the same age? It does seem likely, however, that Edward is one of Samuel's older sons.
Note that neither Hayden Smith, nor any of the younger Smiths discussed above -- Milford, Oliver, David, and Charles M. -- appear in this census. If Emaranda Rowles was born about 1817 or 1818, and was therefore about 16 or 17 when she married Hayden Smith in 1834 in Fountain County, Indiana, it's almost certain that Hayden Smith was at least the same age or, more likely, a bit older. If he was born about 1815, say, he would have been about 35 in 1850 and could easily have been been an older son of Samuel and Margaret Smith. (In fact, since Margaret would have been 17 at the time of his birth, he would probably have been their eldest son.) In that case, William was the younger brother of Hayden and the brother-in-law of Emaranda Rowles. When he moved to Iowa three or four years later, it would like have been in part because he already a brother there.
I have not been able to identify either Samuel or Margaret in the 1860 census, either in Missouri or anywhere else.
And that's where this problem now stands. It's all particularly vexing since this is my paternal great-great-grandfather and there's a natural tendency to want to extend one's "surname" line. Also, I have only three lines which I have not yet been able to extend back before 1800 (one of which descends from an Irish Catholic immigrant, which raises a whole other bunch of well-known problems), and I would really, really like to solve this one!
I may have spent too many hours going over and over the data; perhaps there's something I've stared at too long to be able to see it. If anyone who reads this has any ideas or thoughts -- or if you are also connected to this Smith/Rowles line and have information about any of these folks that I don't -- please let me hear from you. (And thanks for reading this!) --mksmith