I also noted that "I could not find census records for 1920 and 1940."
1) In comments, reader Eddie Black did find 1920 and 1940 U.S. census records, and demonstrated that I had not tried as hard as I should have to find them. Here is what Eddie said:
"I know this isn’t going to help you with your DNA project, but I took note that you were not able to find William Henry Trump and Susan Dill in the 1920 and 1940 census and thought I would give it a try. The 1014 W. Hortter Street is an address that I have passed many times and that made the challenge interesting to me.
"The first thing that I did was look for all of the documents for the Trumps that contained an address.
1911 PA Death Cert - J. D. Trump 515 Westview Ave
1917 PA Death Cert - W. Trump Jr 715 W. Hortter Street
1927 Passenger List 1014 W. Hortter Street
1929 Passenger List 1014 W. Hortter Street
1930 Fed Census 1014 W. Hortter Street
1932 PA Death Cert - M. Trump 1014 W. Hortter Street
1933 PA Death Cert - E. Trump 1014 W. Hortter Street
1935 PHL City Directory 1014 Upsal Street
1942 WWII Draft Registration Alden Park Manor Apts, 5500 Wissahickon Ave
1956 PA Death Cert - W. Trump 8606 Brierdale Rd
1958 PA Death Cert - S. Trump 8606 Brierdale Rd
"In 1920 I found William and Susan Frump at 715 W. Hortter Street.
"In 1940 I found William and Susan? Trumb at 1014 W. Hortter Street.
"The 1935 Philadelphia City Directory had the correct house number but Upsal is a parallel street one block south of Hortter.
"The 1920 census would seem to indicate that William and Susan didn't have any surviving children."
My thanks to Eddie for following up here, and finding the answers for me. His solution, by finding the records by narrowing down the search to addresses, is a very useful tool when you know the addresses from personal knowledge or other records, such as those on Eddie's list. I'm glad that he took the time and had the interest to help me out, and I learned another lesson, and perhaps my readers will absorb the lesson too.
2) In retrospect, I did not do a good job searching the 1920 and 1940 census records on Ancestry.
* If I had searched for "*ump" in the 1920 census, I would have found William H. and Sussan Frump (as indexed):
However, had I searched for "susan" as exact, or born in "New Jersey," I might not have found them.
* If I had searched for "tru*" in the 1940 census, I would have found William and Juska Trumb (indexed as William and Sophia Truont on Ancestry.com:
However, it was not easy because the Ancestry match listed the wife of William as Sophia and I'm not sure I would have clicked on it.
I noted that "Juska" was born in Ohio, so I wondered if that was really Susan (Dill) Trump. I think it is because of the address - that is too much of a coincidence based on Eddie list of addresses.
I have no clue how the enumerator made "Truont" out of "Trumb" or "Sophia" out of "Juska" whatever that means.
3) I wondered how the other record providers indexed that 1940 U.S. Census entry:
* FamilySearch indexed them as William Trumb and Ivaka Trumb. Ivaka is a reasonable transliteration of the record.
* MyHeritage indexed them as William Trumb and Inoka* Trumb. That also is a reasonable transliteration of the record.
* Findmypast indexed them as William Trumb and Ivaka Trumb. That's reasonable too.
The crazy thing is that Susan was the one who provided the information, according to the census image. Was Ivaka or Inoka or Juska a pet name that she had? Or something she made up to confuse future generations of genealogists? Strange, eh?
4) As I always tell my classes - the enumerator didn't have to get the details (name, age, sex, etc) correct, they got paid by the number of names they enumerated.
So, all's well that ends well. No blog post is totally useless - it can always be used as a bad example!
Copyright (c) 2017, Randall J. Seaver