Thursday, September 17, 2015

Mitchell Station -- Part 2

In the last post, Guest Blogger SteamCarriage mentioned an odd artifact found in the fields of Jim Mitchell's Woodside Farm southwest of Hockessin, with the date of 1892 on it. He gave us an excellent background on the creation and evolution of Delaware's (and Mill Creek Hundred's) unique, curved northern border. In this post, SteamCarriage will delve deeper into the events of 1892, and explain exactly how this MCH hill was used. Now, the exciting conclusion of  Mitchell Station...

--Researched and written by SteamCarriage

Stone found in the fields of Woodside
United States Coast and Geodetic Survey
It took nearly 40 more years before another Commission sought out the United States Coast and Geodetic Survey (an early version of the US Geological Survey) to resolve the matter.  The Survey appointed William Chandler Hodgkins to lead the project.  The United States Coast and Geodetic Survey published a very detailed report titled “A Historical Account of the Boundary Line between the States of Pennsylvania and Delaware” by W.C. Hodgkins, dated December 1, 1893.   The 52-page report is available online through Google Books and well worth the read for anyone wanting to know the particulars of the survey.  It is part of the Report of the Superintendent of the US Coast and Geodetic Survey for the fiscal year ending June 30, 1893 (in two parts).

After initial field work, Hodgkins discovered two circles would best define the northern border between Delaware and Pennsylvania and he proposed a solution to the Commission.  They accepted Hodgkins’ analysis and he completed the official boundary survey in 1892-1893 defining the eastern arc boundary of Delaware and Pennsylvania along with the intersection of the Delaware-Maryland border.  The Wedge became part of Delaware and the Delaware-Pennsylvania arc border is a complex arc defined by two different radius arcs with neither center point at the New Castle Court House (but very close).

William Hodgkins had access to the latest equipment and techniques from which to perform his analysis.  His first step was to locate existing markers and survey points and determine their accuracy, etc.  His primary instrument was a repeating theodolite with an 8-inch horizontal scale reading to five seconds of arc.  In conjunction with the theodolite was a heliotrope that used the sun’s rays reflecting off mirrors to form a focused beam of light that could be seen over great distances.  They also used carbide fueled acetylene beacons at night when the air was calm for surveying longer distances.  In order to use the theodolite and heliotrope it was often necessary to stand poles vertically and even build tall scaffolds in order to raise the instruments above tree tops for unobstructed viewing of targets.

Photos of Acetylene Beacon and Theodolite of the era,
 courtesy NOAA (National Geodetic Survey)

With his commissioning on March 8, 1892 W.C. Hodgkins determined the six steps he would take;
Determine the following points which would become the bases for determining the final border;
   o   An official point at the New Castle Courthouse
   o   The Mason-Dixon point for the northeast corner of Maryland
   o   The 12-arc point on the Delaware River
   o   As many points as determined in 1701 still existing
Mapping of these points and determination of various applicable state boundary arcs
Determination by United States Coast and Geodetic Survey commissioners the official arc(s) to be used
Actually survey and placement of preliminary markers for the state boundary line
Review of the surveyed state boundary line by the United States Coast and Geodetic Survey Commissioners to resolve any discrepancies
The permanent marking of the line every half mile with approved stone obelisks

William Hodgkins’ team immediately began scouting out existing boundary markers and then set to work determining their geographical coordinates.  To determine a point’s coordinate, triangulation would be used and the team sought out several key locations from which to determine the actual boundary.  These are shown on the drawing from the survey report and include; I – the Mason-Dixon determined northwestern corner of Maryland; K – a well-defined existing marker on the boundaries of Kennett and Pennsbury townships in Chester County, PA; C – a hickory tree marking the boundaries of Concord and Bethel townships in Delaware County, PA; R – the “ruins”, a location on the Delaware River of a former house or windmill that had been the starting point for previous border surveys.

With his data in hand Hodgkins proposed that the northern arc border between Delaware and Pennsylvania be an arc defined by segments of two arcs.  Based on his analysis he found the eastern part of the DE-PA border (basically the boundary common to Delaware County, PA and New Castle County, DE) from a point known as Kennett-Pennsbury east to “the ruins” on the Delaware River to be reasonably accurate.  An arc of radius 12.82 miles drawn from a point actually located in the Delaware River southeast of the New Castle Courthouse would provide an accurate definition of the border from Kennett-Pennsbury to the Delaware River.

For the western part of the boundary (basically the boundary common to Chester County, PA and New Castle County, DE), an arc of radius 11.58 miles centered on a point northwest of the courthouse would define the boundary between Kennett-Pennsbury and the Mason-Dixon Line going west.  Hodgkins’ report of 1893 details the geometry he proposed to finally define the Maryland-Delaware-Pennsylvania boundaries.

1892 Survey Points

With approval from the Commission Hodgkins returned to place the actual Delaware-Pennsylvania boundary markers at ½ mile intervals.  In order to do this it was necessary to erect stations on either side of the boundary by designating major triangulation points.  These included Hamorton, Londonderry, White, and Whiteman in PA, Centerville, Meetinghouse Hill, and Mitchell’s Farm in DE, and Grays Hill in MD.  From these primary points, secondary triangulation points were determined from which the actual boundary points could be placed.

From known stations at locations such as Centerville and Meeting House Hill, DE, Londonderry, PA, and Gray’s Hill, MD other stations were defined including Mitchell Station on Woodside Farm, DE.  Mitchell Station would be defined from Meeting House Hill and Walnut Stations.  Mitchell Station would then be used to determine Hoopes, Foulk, and Southwood stations in PA as well as O’Neill and Stephen stations in DE.  Each of these stations would then be used to determine the precise locations of each of the 1/2–mile markers between mileposts 2 and 6.

Mitchell Station, located on the Mitchell’s Woodside Farm was one of the major points used in determining at least seven other triangulation points.  Located on a high knoll on the farm, a wooden tower was erected similar to the one pictured below.  This would have been a substantial station having been used for both Heliotrope, carbide beacon, and Theodolite operations.  Mr. Mitchell has noted that stories have been passed down through the generations that a tall wooden tower was erected by the survey team and that many nights the surveyors were there all night tending the very bright beacon.  As night air is typically more still and uniform in temperature, the beacon was seen as stationary at great distances instead of scintillating due to air density caused refraction effects.

Examining the 1892 Survey’s Data
In reading William Hodgkins “A Historical Account of the Boundary Line between the States of Pennsylvania and Delaware” we got to wondering just how accurate the data might have been as compared to modern online maps that are available.  With survey triangulation point coordinates, as well as the coordinates for each of the final boundary stations was entered into an Excel spreadsheet that could be imported into Google Visualizer we started drawing some maps.

One of the first maps to draw simply plotted the survey points in Google Maps, on a terrain map.  One of the features of GPS Visualizer is that it can draw tracks for selected groups of points so we let it do this for several of the point groupings.  On the map you’ll see the three 12-mile arcs leading to the boundary disputes.  The series of triangles and squares mark the present day border of Delaware-Pennsylvania and you’ll note the markers fall on the either the light or dark tan arcs depending if your east (triangles) or west (squares) of the Kennett-Pennsbury marker noted on the map.  The red arc, for reference, is the 12-mile point from the steeple on the New Castle County Courthouse and represents what the boundary “should” have been had today’s surveying equipment and techniques been available back in 1701.  Around the New Castle Court House’s red star are the arc center point locations, marked by light/dark tan stars, for the other two arcs as found by Hodgkins’ team.

If we plot the points and arcs on a terrain map with state borders and some color shading for elevations, the Wedge clearly stands out at the middle left.  With the red arc marking 12-miles from the courthouse steeple, it shows how the dispute for where the Delaware-Maryland, and Delaware-Pennsylvania boundaries were before the survey.  Choose the arc point at the Maryland east-west boundary with Pennsylvania and the northeast corner of Maryland is at Whiteman.  Use the tangent point of the arc and a north-south boundary line for Maryland-Delaware and the northeast corner of Maryland is its modern day location.

Also note that the Delaware-Maryland border from the northeast corner of Maryland to the Tangent Point is exactly north-south, however that south of the Tangent Point the Delaware-Maryland border, while a straight line, is not true north-south but at an angle.  Many maps show this Delaware-Maryland boundary line as straight along the entire western side of Delaware and it is not.

It is also interesting to note how the modern map eastern border for Delaware, along the New Jersey shore of the Delaware River, followed the Delaware eastern arc line to jump back to the center of the Delaware River as we cross from Delaware into Pennsylvania.

One of the features of Google Earth is to be able to zoom around the countryside and take a low-level bird’s eye view.  If we fly from the Atlantic Ocean coming northwesterly over New Jersey we can get an idea of relative heights of the major stations.  For example, from New Castle Courthouse they had to be able to see Gray’s Hill, Meeting House Hill and Centerville.  Seeing those allowed them to coordinate with Londonderry and White.  They could then set out major points such as Walnut, Mitchell and Stephen, etc. You’ll see those major points in the map below standing above those of the actual boundary stones forming the border arc.

There has been much written over the years about how the various borders of Delaware, Maryland, Pennsylvania, and New Jersey were established.  Unfortunately some accountings contain errors.  Having access to the United States Coast and Geodetic Survey’s report “A Historical Account of the Boundary Line between the States of Pennsylvania and Delaware” by W.C. Hodgkins, the details of the final determination of the 12-mile arc boundary are been well documented.


  1. Wow, well done SteamCarriage. I'll read this a few times. The northern boundary of Delaware is not one arc, but a combination of two. I never knew the final reconciliation came in 1892, I'll bet few people do.

    Have you looked for (or found reference to) any other artifacts from that survey, maybe from other locations?

    Thanks for sharing.

  2. Agreed, Walt. If you get nothing else out of these posts, just the realization of how complicated it all has been is fascinating. "It's a 12 mile circle from New Castle" doesn't nearly cover it.

    I was thinking about other locations, too. The big one seems to have been Meeting House Hill, which Scharf also mentions being used in the 1850s for some sort of survey of the Atlantic coast. Probably used about the same spot, somewhere off of Poly Drummond Hill Road. Love to know if there's another stone buried somewhere in someone's backyard.

  3. Walt C -- Thanks. The actual 1/2-mile spaced arc stones are mostly still in place as far and I know (haven't personally checked each and every one). If you click on the maps I did with the squares & triangles marking the boundary, you'll see a couple of the squares/triangles are black. Those markers on the boundary are documented to be "missing".

    When I was preparing the article I did overlay other maps, like developments, on Google Maps of the survey data and the various points of the survey used to determine the actual boundary stone locations were in housing developments and what not. I think it is highly unlikely that any of the stones are still buried outside of the major starting locations that were well designated at the start of the survey.

    The stone on Mitchell's farm had been buried since 1892 and only came to the surface after the field was plowed (hence the scar from center left to top right). Its unknown how many times the field had been plowed in over 120 years but it wasn't that far below the surface according to Jim Mitchell. I rather think it was a miracle it remained hidden all the years that it did.

  4. I know I have some pictures of one of the 1892 1/2 mile markers from up around Corner Ketch. Next time I'm on my computer at home I'll post it so people can see what they look like.