The I-10/I-110 interchange in Los Angeles - a postcard favorite?

[TheStranger]

Going through Flickr and searching for shots of the LA freeway system on postcards, one would think that the world-famous (and still trend-setting) Four-Level would be the primary source of photos for the tourists to mail back home.

And while it is true that that interchange and the freeways that connect it (the Harbor and Hollywood freeways and the Arroyo Seco Parkway, at that point the Pasadena Freeway) are prominently featured in photos, the slightly less vintage left-exit cloverstack where the Santa Monica and Harbor Freeways meet up seems to have its own postcards more often than I expected!

More often than not, they received the "Man, dig those crazy Los Angeles freeways!" captions, which to me is surprising - the Four-Level at least was a trendsetter, while the most "crazy" interchange in the area is the East Los Angeles Interchange just 2 miles to the east.

http://www.flickr.com/photos/wilsonism/188164365/
http://www.flickr.com/photos/bastian/104582811/
http://www.flickr.com/photos/rohtola/3256758845/
http://www.flickr.com/photos/striderv/2701531134/
http://www.flickr.com/photos/striderv/4578600556/

[J N Winkler]

Years ago, while browsing an old-books aggregator site for research-related books (either ABEBooks.com or Bookfinder.co.uk), I found an advertisement for a vintage set of construction plans for the I-10/I-110 interchange.  (I didn't buy it--for one thing, they wanted too much--it was in the hundreds of dollars--and for another, I was really interested in the Four Level construction plans, which I found several years later in Los Angeles' Engineering Vault.)

[TheStranger]

Years ago, while browsing an old-books aggregator site for research-related books (either ABEBooks.com or Bookfinder.co.uk), I found an advertisement for a vintage set of construction plans for the I-10/I-110 interchange.  (I didn't buy it--for one thing, they wanted too much--it was in the hundreds of dollars--and for another, I was really interested in the Four Level construction plans, which I found several years later in Los Angeles' Engineering Vault.)

Was the interchange planned before I-10 was assigned on the Santa Monica Freeway? (I know the planned Olympic Parkway was State Route 26 - originally assigned in 1934 as bear State Route 6 before US 6 arrived in the state - but its route does not completely parallel the eventual Santa Monica alignment).

For being a cloverstack, it manages to take up more right of way than the Four-Level (albeit using higher-speed ramps) AND includes the strange left-exit/left-entry from east I-10 to north I-110!  To me that has always struck me as the oddest part of the design, especially when the Four-Level accomplished semi-directionality rather elegantly just 2 miles up the Harbor Freeway.

I guess the planners expected more through traffic from Santa Monica (and by extension, LAX) to only be heading to downtown, rather than to the East Los Angeles Interchange? 

[J N Winkler]

I can't really answer your questions as framed, because I am not an expert on the history of the LA freeway network.  However, this plan sheet (almost unreadable due to microfilm "snow") implies that the segment of I-10 running westward from the East Los Angeles Interchange postdated that interchange, which itself did not begin to be constructed until 1959 (BTW, the mainline Santa Ana Freeway already ran through the projected East Los Angeles Interchange complex, which to my eye accounts for some of the awkward ramp alignments):

http://engvault.lacity.org/images/tif/tape001/dvd0008/images000055000/la000054113.tif

Parts of the Olympic Parkway were likely built as the Olympic Parkway.  The Engineering Vault has scanned microfilm for an Olympic Parkway/Santa Ana Parkway interchange (probably I-5/I-10):

http://engvault.lacity.org/images/tif/tape001/dvd0003/images000025000/la000024247.tif

Later, in 1960, Caltrans released for construction a set of signing plans which were part of a larger turnkey contract for construction on Route 173 (what we now know as I-10).  These called for SR 26 shields to be covered up in favor of I-10 shields.  The federal-aid project number for this job was I-010-1(15)16 but if you want to dig it up through the Engineering Vault, I would recommend doing a search for "Santa Monica Frwy" and zeroing in on the relevant stuff by project limits (8th St. to 0.3 mile west of Hooper Ave.).  Signing plans are here:

http://engvault.lacity.org/vault/public/vault_search_result.cfm?index_no=D-15543

In regard to the expansiveness of the Four Level, I believe it is unusual in several respects other than being the first four-level Maltese cross stack.  First, it is probably the only stack in the US whose bridge components are unified under a single NBI number.  This is because the Four Level structure itself is just one bridge with multiple levels.  Second, I am all but certain it holds the record for smallest number of plan sheets involved in construction.  The plans set for the bridge structure itself has just 90 sheets, and approach roadways add a few dozen sheets.  The nearest runner-up I have found is a stack in the Cleveland area done under a turnkey contract for which the plans set had about 400 sheets.  A modern stack can be done in as few as 1600 sheets (NTTA) but 2000 sheets on up is more usual.  The High Five and the two Katy Freeway stacks were over 4000 sheets each.  Third, because the Four Level was planned as a drop-in replacement for a cloverleaf between two hills where it would have been impracticably expensive to build an actual cloverleaf, it has probably the lowest ramp operating speeds of any stack anywhere, and as a result has little additional traffic capacity over and above a comparably dimensioned cloverleaf.

[TheStranger]

I can't really answer your questions as framed, because I am not an expert on the history of the LA freeway network.  However, this plan sheet (almost unreadable due to microfilm "snow") implies that the segment of I-10 running eastward from the East Los Angeles Interchange postdated that interchange, which itself did not begin to be constructed until 1959 (BTW, the mainline Santa Ana Freeway already ran through the projected East Los Angeles Interchange complex, which to my eye accounts for some of the awkward ramp alignments):

http://engvault.lacity.org/images/tif/tape001/dvd0008/images000055000/la000054113.tif

IIRC, all of the Santa Ana Freeway as we know it today (albeit not as wide in parts as it has been made over the years) existed as US 101 prior to the start of the construction of the East Los Angeles Interchange, I'd even hazard a guess that it was the first or second segment of today's Interstate 5 built anywhere (along with San Diego's Montgomery Freeway) as it predates the Interstate system.

Although considering the changes made to the North Sacramento Freeway to accomodate what was then US 99E/I-80 in the late 1950s/early 1960s, I'm surprised Division of Highways didn't try to simplify the configuration here (i.e. have the Golden State Freeway end at the San Bernardino Split instead of the eastern terminus of the Santa Monica Freeway).

Parts of the Olympic Parkway were likely built as the Olympic Parkway.  The Engineering Vault has scanned microfilm for an Olympic Parkway/Santa Ana Parkway interchange (probably I-5/I-10):

http://engvault.lacity.org/images/tif/tape001/dvd0003/images000025000/la000024247.tif

Later, in 1960, Caltrans released for construction a set of signing plans which were part of a larger turnkey contract for construction on Route 173 (what we now know as I-10).  These called for SR 26 shields to be covered up in favor of I-10 shields.  The federal-aid project number for this job was I-010-1(15)16 but if you want to dig it up through the Engineering Vault, I would recommend doing a search for "Santa Monica Frwy" and zeroing in on the relevant stuff by project limits (8th St. to 0.3 mile west of Hooper Ave.).  Signing plans are here:

http://engvault.lacity.org/vault/public/vault_search_result.cfm?index_no=D-15543

Looking at Mike Ballard's site on the East Los Angeles Interchange, it appears that the Hooper-to-8th segment of the Santa Monica Freeway didn't yet exist in 1959 - suggesting that this was more the case of reworking plans yet to be implemented, rather than a resigning project (in this segment) that consisted of an in-the-field shield relpacement:
http://www.scvresources.com/highways/east_los_angeles_interchange.htm

In regard to the expansiveness of the Four Level, I believe it is unusual in several respects other than being the first four-level Maltese cross stack.  First, it is probably the only stack in the US whose bridge components are unified under a single NBI number.  This is because the Four Level structure itself is just one bridge with multiple levels.  Second, I am all but certain it holds the record for smallest number of plan sheets involved in construction.  The plans set for the bridge structure itself has just 90 sheets, and approach roadways add a few dozen sheets.  The nearest runner-up I have found is a stack in the Cleveland area done under a turnkey contract for which the plans set had about 400 sheets.  A modern stack can be done in as few as 1600 sheets (NTTA) but 2000 sheets on up is more usual.  The High Five and the two Katy Freeway stacks were over 4000 sheets each.  Third, because the Four Level was planned as a drop-in replacement for a cloverleaf between two hills where it would have been impracticably expensive to build an actual cloverleaf, it has probably the lowest ramp operating speeds of any stack anywhere, and as a result has little additional traffic capacity over and above a comparably dimensioned cloverleaf.

Does the general brevity of plan sheets derive from how old the interchange is?  It also seems to me that no new stacks were built nationwide until the 1960s.

While a tightly-built stack like this one does not offer much more capacity than a cloverleaf, it at the very least eliminates weaving - and there certainly was more right of way available at the Santa Monica/Harbor junction to build a faster, higher-capacity stack as opposed to what ended up being constructed.  Having said that, the age of that 110/10 interchange could be such as to have been before the larger true stack interchanges came into vogue.

[Alps]

The one I'm familiar with is the I-105/110 interchange just off LAX, whatever it may be named, because it's visible from the airplane right as you land.  It's plenty large and crazy (and postcard-worthy) to me.

[TheStranger]

Just found yet another postcard showing this junction, albeit this time looking towards the west:
http://www.flickr.com/photos/bison132/4569653540/

Probably the most interesting one so far.

http://www.flickr.com/photos/jakegordon/151743901/ provides another angle, looking slghtly to the northwest from the site of Metropolitan Junior College.

[J N Winkler]

Parts of the Olympic Parkway were likely built as the Olympic Parkway.  The Engineering Vault has scanned microfilm for an Olympic Parkway/Santa Ana Parkway interchange (probably I-5/I-10):

http://engvault.lacity.org/images/tif/tape001/dvd0003/images000025000/la000024247.tif

Later, in 1960, Caltrans released for construction a set of signing plans which were part of a larger turnkey contract for construction on Route 173 (what we now know as I-10).  These called for SR 26 shields to be covered up in favor of I-10 shields.  The federal-aid project number for this job was I-010-1(15)16 but if you want to dig it up through the Engineering Vault, I would recommend doing a search for "Santa Monica Frwy" and zeroing in on the relevant stuff by project limits (8th St. to 0.3 mile west of Hooper Ave.).  Signing plans are here:

http://engvault.lacity.org/vault/public/vault_search_result.cfm?index_no=D-15543

Looking at Mike Ballard's site on the East Los Angeles Interchange, it appears that the Hooper-to-8th segment of the Santa Monica Freeway didn't yet exist in 1959 - suggesting that this was more the case of reworking plans yet to be implemented, rather than a resigning project (in this segment) that consisted of an in-the-field shield replacement:

http://www.scvresources.com/highways/east_los_angeles_interchange.htm

I would have to download the complete plans set (which is not straightforward from the Engineering Vault site since it is chopped up into different parts by functional discipline) to get an idea of the scope of work, but I don't think we are dealing with a "construction revisions" situation.  For one thing, when changes are made to a project during construction, it is customary for the old plan sheets to be left in the plans set but with cancel bars through them, and that is not the case here.  I think the SR 26 shields relate instead to a wye connector between the Santa Ana Parkway and the projected Olympic Parkway which was probably built in the late 1940's and then removed when the East Los Angeles Interchange was built.  If this hypothesis is correct, the roadway plans for I-10 8th St. to Hooper or for the first East Los Angeles Interchange contract (which was begun in 1959 and has an I-005-XXXX project number which escapes me right now) should contain sheets providing for the obliteration of the Olympic Parkway wye connection.  Google Maps shows what appears to be an abandoned grade in an approximately correct location.

In regard to the expansiveness of the Four Level, I believe it is unusual in several respects other than being the first four-level Maltese cross stack.  First, it is probably the only stack in the US whose bridge components are unified under a single NBI number.  This is because the Four Level structure itself is just one bridge with multiple levels.  Second, I am all but certain it holds the record for smallest number of plan sheets involved in construction.  The plans set for the bridge structure itself has just 90 sheets, and approach roadways add a few dozen sheets.  The nearest runner-up I have found is a stack in the Cleveland area done under a turnkey contract for which the plans set had about 400 sheets.  A modern stack can be done in as few as 1600 sheets (NTTA) but 2000 sheets on up is more usual.  The High Five and the two Katy Freeway stacks were over 4000 sheets each.  Third, because the Four Level was planned as a drop-in replacement for a cloverleaf between two hills where it would have been impracticably expensive to build an actual cloverleaf, it has probably the lowest ramp operating speeds of any stack anywhere, and as a result has little additional traffic capacity over and above a comparably dimensioned cloverleaf.

Does the general brevity of plan sheets derive from how old the interchange is?  It also seems to me that no new stacks were built nationwide until the 1960s.

I can't be confident that no Maltese cross stacks were built in Detroit or Flint during the 1950's.  One of the Ohio stacks may have been started in the 1950's too.  I would hazard that Maltese Cross Stack No. 2 in California was the I-10/I-405 interchange in Santa Monica (1965, if memory serves).

It was definitely normal for plan sets to be bare-bones back in the 1940's and 1950's.  A simple freeway overpass in California in the late 1940's typically ran to 10-13 sheets while today in most states, and indeed in California by the late 1950's, it would run to more like 20-25 sheets.  But I can't imagine the Four Level running to more than about 300 sheets for bridges alone under modern drafting standards.  It is a single composite structure, and the design is conceptually simple although there is much need to detail grades, superelevation, and disposition of reinforcement on the "wishbone" segment between the US 101 and SR 110 levels where one ramp on a rising grade goes past another on a falling grade.  There is relatively little centerline length on the ramps that is actually part of a bridge structure--modern stacks (especially in flat country in Texas) tend to have a lot of length on bridge deck on the connectors and that factor alone requires the use of multiple sheets just to show a complete elevation along the PGL at an acceptably large scale.

While a tightly-built stack like this one does not offer much more capacity than a cloverleaf, it at the very least eliminates weaving - and there certainly was more right of way available at the Santa Monica/Harbor junction to build a faster, higher-capacity stack as opposed to what ended up being constructed.  Having said that, the age of that 110/10 interchange could be such as to have been before the larger true stack interchanges came into vogue.

It is certainly possible that the Four Level had no immediate followers because it was considered that its design was made possible only by fortuitous topography and that techniques for traffic estimation had not progressed to the point where a justification for stacks could be shown under less favorable topographical conditions.  But this question is hard to answer without a design study report for the I-10/I-110 interchange.  I assume one was prepared but has been either archived or lost.

[TheStranger]

I would have to download the complete plans set (which is not straightforward from the Engineering Vault site since it is chopped up into different parts by functional discipline) to get an idea of the scope of work, but I don't think we are dealing with a "construction revisions" situation.  For one thing, when changes are made to a project during construction, it is customary for the old plan sheets to be left in the plans set but with cancel bars through them, and that is not the case here.  I think the SR 26 shields relate instead to a wye connector between the Santa Ana Parkway and the projected Olympic Parkway which was probably built in the late 1940's and then removed when the East Los Angeles Interchange was built.  If this hypothesis is correct, the roadway plans for I-10 8th St. to Hooper or for the first East Los Angeles Interchange contract (which was begun in 1959 and has an I-005-XXXX project number which escapes me right now) should contain sheets providing for the obliteration of the Olympic Parkway wye connection.  Google Maps shows what appears to be an abandoned grade in an approximately correct location.

I think the abandoned interchange is shown in Ballard's pages as well, albeit from what I remember reading there, it never was hooked up to the Olympic Parkway/Santa Monica Freeway (as the East Los Angeles Interchange ramps completely supplanted it).

I can't be confident that no Maltese cross stacks were built in Detroit or Flint during the 1950's.  One of the Ohio stacks may have been started in the 1950's too.  I would hazard that Maltese Cross Stack No. 2 in California was the I-10/I-405 interchange in Santa Monica (1965, if memory serves).

Detroit's most notable 1950s interchange is the Lodge/Edsel Ford junction - similar to a stack, but not semi-directional - which Wikipedia claims as the first full freeway-to-freeway junction in the US when it opened in 1953 (but I'm not sure this is entirely accurate).  Looking briefly at Google Maps, there doesn't appear to be any other true symmetrical stack interchanges, though the Jeffries/Ford junction (94 and 96) comes close, albeit a 1970s construct that is slightly assymetrical.

In Flint, the only stack is 475/69; 475 was first built in 1973 though so it postdates several other stacks.

Continuing the Google Maps examinations, looking at Ohio now, the 490/77 stack in Cleveland seems to be a much more recent addition - 490 wasn't completed until 1990.  490/90 may have come earlier but has one ramp movement missing, though it is functionally a true semi-directional stack.  480/77 I think was a 1960s or 1970s build.

While a tightly-built stack like this one does not offer much more capacity than a cloverleaf, it at the very least eliminates weaving - and there certainly was more right of way available at the Santa Monica/Harbor junction to build a faster, higher-capacity stack as opposed to what ended up being constructed.  Having said that, the age of that 110/10 interchange could be such as to have been before the larger true stack interchanges came into vogue.

It is certainly possible that the Four Level had no immediate followers because it was considered that its design was made possible only by fortuitous topography and that techniques for traffic estimation had not progressed to the point where a justification for stacks could be shown under less favorable topographical conditions.  But this question is hard to answer without a design study report for the I-10/I-110 interchange.  I assume one was prepared but has been either archived or lost.

If there's an assessment/analysis out there for that particular junction, I'd love to see it - the 10/110 interchange doesn't really represent a design that was used elsewhere.  At the very least, the usage of the left exit/left entrance coming from the west half of the Santa Monica Freeway seems unusual, as it doesn't appear to be driven by topography (which is why I figure the design philosophy had to do with funneling traffic to downtown from Santa Monica).  I don't think any new CalTrans interchange projects, especially in metro Los Angeles, since have involved a freeway "exiting off of itself" in a full mainline switch as part of the original design (I-80 does this now, but only due to numeric redesignation in Sacrament)...though the Route 94/Route 15 (future I-15) junction in San Diego may postdate this slightly, and is slated for upgrade to semi-directionality some time this decade. 

[J N Winkler]

Detroit's most notable 1950s interchange is the Lodge/Edsel Ford junction - similar to a stack, but not semi-directional - which Wikipedia claims as the first full freeway-to-freeway junction in the US when it opened in 1953 (but I'm not sure this is entirely accurate).  Looking briefly at Google Maps, there doesn't appear to be any other true symmetrical stack interchanges, though the Jeffries/Ford junction (94 and 96) comes close, albeit a 1970s construct that is slightly assymetrical.

According to my stack list, which is not with me now, Michigan has 4 stacks--three in the Detroit area and one in Flint.  This list does not include the Lodge/Edsel interchange, which as you note is not a fully semidirectional interchange.  For Detroit, in addition to I-94/I-96, which I count as a true stack because the ramp layout has the correct topological characteristics, there is I-96/M-39 (late 1960's would be my guess) and I-96/I-75 (probably also late 1960's).

My doubt about the dating of these stacks arises from the use of what appear to be vertical riveted stiffeners on the bridge girders in (at least) I-96/M-39.  The use of these stiffeners seems to be typical of long bridges built in the 1950's, including (IIRC) the now-demolished original Cuyahoga River bridge on the Ohio Turnpike.

Continuing the Google Maps examinations, looking at Ohio now, the 490/77 stack in Cleveland seems to be a much more recent addition - 490 wasn't completed until 1990.  490/90 may have come earlier but has one ramp movement missing, though it is functionally a true semi-directional stack.  480/77 I think was a 1960s or 1970s build.

I-77/I-490 is hoary because it was built with the Willow Freeway (I-77).  I'd say I-77/I-480 was built in the same decade.  I actually have the complete construction plans for both stacks, as well as for the defective stack at I-90/I-490, but not within reach . . .

In regard to the uniqueness of the I-10/I-110 interchange, the US 35/Jefferson Ave. interchange in Dayton, Ohio (whose layout may have changed since, last I checked, US 35 was slated for major reconstruction) has some features in common with I-10/I-110, including a left exit past the cross road which leads to the cross road by means of a loop.  Neither Jefferson Ave. nor SR 48 (which is part of the same general interchange complex) is a freeway, but I think a freeway might have been projected in this corridor and subsequently cancelled.

BTW, Dayton has Ohio's third stack--I-75/US 35.

[TheStranger]

Detroit's most notable 1950s interchange is the Lodge/Edsel Ford junction - similar to a stack, but not semi-directional - which Wikipedia claims as the first full freeway-to-freeway junction in the US when it opened in 1953 (but I'm not sure this is entirely accurate).  Looking briefly at Google Maps, there doesn't appear to be any other true symmetrical stack interchanges, though the Jeffries/Ford junction (94 and 96) comes close, albeit a 1970s construct that is slightly assymetrical.

According to my stack list, which is not with me now, Michigan has 4 stacks--three in the Detroit area and one in Flint.  This list does not include the Lodge/Edsel interchange, which as you note is not a fully semidirectional interchange.  For Detroit, in addition to I-94/I-96, which I count as a true stack because the ramp layout has the correct topological characteristics, there is I-96/M-39 (late 1960's would be my guess) and I-96/I-75 (probably also late 1960's).

Reason I didn't mention the I-96 stacks - Wikipedia notes that the Jeffries was built entirely in the 1970s, starting with the segment from the Fisher Freeway (I-75) to the Ford Freeway (I-94) in 1970.  This would place it five years after California's second stack (405/10) was built.  (Part of me wants to say the third California stack would be today's 24/580/980 junction just east of the MacArthur Maze in Oakland)

The Jeffries/Southfield junction you mention DOES use less right of way than newer stacks, so that makes me wonder if any of the Jeffries was built as M-14 (with the stack coming before the I-96 designation), contrary to what Wiki says.

Continuing the Google Maps examinations, looking at Ohio now, the 490/77 stack in Cleveland seems to be a much more recent addition - 490 wasn't completed until 1990.  490/90 may have come earlier but has one ramp movement missing, though it is functionally a true semi-directional stack.  480/77 I think was a 1960s or 1970s build.

I-77/I-490 is hoary because it was built with the Willow Freeway (I-77).  I'd say I-77/I-480 was built in the same decade.  I actually have the complete construction plans for both stacks, as well as for the defective stack at I-90/I-490, but not within reach . . .

In regard to the uniqueness of the I-10/I-110 interchange, the US 35/Jefferson Ave. interchange in Dayton, Ohio (whose layout may have changed since, last I checked, US 35 was slated for major reconstruction) has some features in common with I-10/I-110, including a left exit past the cross road which leads to the cross road by means of a loop.  Neither Jefferson Ave. nor SR 48 (which is part of the same general interchange complex) is a freeway, but I think a freeway might have been projected in this corridor and subsequently cancelled.

BTW, Dayton has Ohio's third stack--I-75/US 35.

Looking at the the Jefferson St/US 35 interchange, I completely see the idea of a left-exit loop (though it is only from one direction) in play -
http://maps.google.com/maps?f=q&source=s_q&hl=en&geocode=&q=Jefferson+Ave,+Dayton&sll=39.742966,-84.13373&sspn=0.043888,0.104628&ie=UTF8&hq=&hnear=S+Jefferson+St,+Dayton,+Montgomery,+Ohio+45402&ll=39.753805,-84.186022&spn=0.01097,0.026157&z=16

How old is that US 35 freeway?

[J N Winkler]

I am not sure about Dayton US 35.  My guess is late 1950's/early 1960's for the SR 48 interchange complex.  I would have to dig up the title sheet for the plans . . .

[TheStranger]

I am not sure about Dayton US 35.  My guess is late 1950's/early 1960's for the SR 48 interchange complex.  I would have to dig up the title sheet for the plans . . .

I do find it interesting that in both the 10/110 junction and this US 35/SR 48 example, the left-exit flyover is a straight shot to downtown, as if that was the more important destination than points east...

Not counting situations where a route designation leaves a pre-existing mainline (so the Santa Ana Freeway continuing to downtown Los Angeles does not count as an intentional example of I-5 traffic being funneled away from itself towards downtown, due to that whole named route once being all US 101)...are there any other instances in which a junction was built with the mainlines exiting/entering on the right, to funnel traffic to a specific destination like this?

1950s freeway construction in metro Los Angeles did bring a variety of freeway-to-freeway interchange types that hasn't been seen more recently in an age of either cloverstacks or pure stacks - in addition to the East Los Angeles Interchange, the Four-Level, and this 10/110 junction...we had the left-exit-heavy Long Beach Freeway/Santa Ana Freeway junction, the Y-interchange known as the San Bernardino Split, the earliest version of the Newhall Pass interchange (at the time, US 99 and US 6), and the trumpet that linked the Colorado Freeway with the Golden State Freeway - all interchanges that had nothing to do with each other in form.  (The San Bernardino Freeway/Long Beach Freeway junction, originally the linkage of the 60-70-99/15 freeways when constructed, was built in 1960 - just outside of the 1950s era variety, but before that second stack in West Los Angeles in '65.)

As mentioned earlier, the Four-Level originally appears to have been a product of geographic circumstance (which appears to be the motivation for most of the early freeway-to-freeway junction designs too) rather than simply the most ideal design for traffic flow, though the elimination of weaving proved highly influential in the years to come.

Having said all that, I want to hazard a guess that the original late-1930s MacArthur Maze is California's first freeway-to-freeway interchange, albeit the original structures have since been replaced with newer ones added in the late 1950s.