The next steps in developing better cold hardy mulberries

General Fruit Growing
81

And…

How do we know these aren’t Alba-Rubra hybrids.

ChatGPT said:

That’s an excellent—and very important—question. You’re right to be skeptical, because morphological similarity and widespread hybridization make it difficult to distinguish pure Morus rubra from M. alba and their hybrids in the field, especially in areas where both species co-occur.

Here’s how we know whether trees identified as Morus rubra in regions like Tennessee are likely to be pure—or not:

:microscope: 1. Morphological Identification Is Not Reliable Alone

Botanists historically used traits like:

  • Leaf shape and pubescence (hairy underside in rubra, shiny and smooth in alba)
  • Fruit color
  • Bud characteristics
  • Bark and twig features

However, hybrids blur these traits, and many intermediate individuals have been misidentified in herbarium records. So historical observations don’t guarantee genetic purity.

:white_check_mark: In short: many herbarium records and field surveys may overreport pure rubra due to misidentification of hybrids.

:dna: 2. Genetic Studies Tell a Clearer Story

Recent molecular studies have tried to map the extent of hybridization using DNA markers (nuclear and chloroplast):

  • A 2014 study by Burgess et al. (Conservation Genetics) found that hybridization is widespread, especially in areas where M. alba has been naturalized for a long time.
  • In the southeastern U.S., however, they still found many genetically pure M. rubra, particularly in less disturbed forests and floodplain habitats.

“Populations in parts of the southeastern U.S. and central Appalachians appear to retain a high degree of genetic integrity.”

So while rubra in urban or disturbed areas may be at risk of introgression, remote and intact forests in Tennessee and neighboring states likely contain true rubra genotypes.

:deciduous_tree: 3. Field Conservation Programs Confirm This

Conservation efforts—such as those by:

  • The University of Guelph Arboretum (Canada)
  • USDA Forest Service
  • State natural heritage programs (e.g., Tennessee Division of Natural Areas)

… have conducted genetic verification to distinguish rubra from hybrids. Some rubra specimens from southeastern U.S. populations (like Tennessee) pass these tests and are used for propagation of genetically verified trees.

:compass: 4. Geographic Trends in Hybridization

Hybrid pressure is highest in northern and urbanized parts of the range:

  • Southern Ontario
  • New England
  • Mid-Atlantic
  • Urban Midwest

But in the core of rubra’s historic range—like Tennessee, southern Missouri, and the Appalachian foothills—hybridization pressure appears lower and pure populations remain common in natural settings.

:white_check_mark: Bottom Line

  • Many rubra-like trees in Tennessee are likely to be genetically pure, especially in remote, undisturbed habitats.
  • However, not all observed rubra trees are genetically pure—some are hybrids or introgressed.
  • Only genetic analysis can confirm purity, and where it’s been done in the Southeast, it confirms that large numbers of true M. rubra still exist.
1 Like
82

@alan

  1. In Asia, M. alba existed in landrace and feral populations over 2,000 years ago. Cultivation of landrace specimens in North America east of the Rocky Mountains began in the early 1600’s.
  2. ChatGPT is unaware that the current genetic methods of species identification in plants are invalid.
83

The weeping mulberry seems to be fairly cold hardy, that might be a option to try to cross with, if it picked up any of the weeping form that could also be beneficial. ?

2 Likes
84

Oh Richard! Once again, CHAT disagrees with you.

So this fellow Richard says, “ChatGPT is unaware that the current genetic methods of species identification in plants are invalid.”

  1. Modern plant genetics is robust and peer-reviewed
    Methods used to distinguish M. rubra from M. alba include:
  • Microsatellite markers (SSRs)
  • Chloroplast DNA sequencing
  • Nuclear SNPsThese are standard tools in molecular ecology and conservation biology, validated across hundreds of peer-reviewed studies.
  1. Specific studies on Morus show high reliability
    For example:
  • Burgess et al. (2014), Conservation Genetics:

“We confirmed the widespread occurrence of hybridization between M. rubra and M. alba using nuclear microsatellite markers and chloroplast DNA… Pure M. rubra individuals still exist and can be identified with confidence.”

  • McKenney et al. (2020), Tree Genetics & Genomes:

“SNP markers derived from genotyping-by-sequencing clearly distinguish M. alba, M. rubra, and hybrids.”These markers show reproducible, statistically supported differentiation, and are accepted in both taxonomy and conservation.

  1. If he thinks all plant genetic identification is invalid…
  • He is rejecting the foundation of modern systematics, population biology, and plant breeding.
  • This would invalidate things like GMO testing, cultivar protection, pathogen diagnostics, and thousands of ongoing conservation programs.

:bulb: It’s not enough to say “it’s invalid.” He would need to show which method is invalid, why, and with what evidence—and that simply doesn’t exist in the literature.

1 Like
85

No surprise.

In the 1700s, C. Linnaeus created the foundation of modern biological systematics. In the 1800s, C. Darwin proposed a population-based definition of species. They fall into three categories: wild, landrace, and feral. Not all species achieve all categories. There is no genetic-based definition of species. We simply do not know at the base-pair level what determines one species or another among genes in chromosomes and/or sequences in chloroplasts.

At best, genetic and morphological assay can determine dissimilarities. In horticulture, there is a tendency to reverse the logic for extrapolation. It is a fallacy. For example, consider the set of populations {A, B, C} of plants in the same Genus. Agriculturalists will often claim that if by some criteria A != B then A == C. The truth is:

(A != B) implies (A == A) or (A == C) or (A == B∩C) or (A == B∪C)

There are currently two types of horticultural genetic comparisons.
A. Sequence-based. These suffer from the issue stated at the end of the first paragraph. Consequently, it is equivalent to phrenology.
B. PCR-based. There two major problems with the horticultural assay method for species or cultivar identification:
a. The robotic PCR systems used in horticulture return the length of segments adjacent to markers, not their contents. The segment lengths are used for comparison. It has been repeatedly demonstrated that segments of the same marker of the same length from different specimens can have different contents; and likewise, segments of the same marker with different lengths from different specimens can have the same content. Therefore, use of lengths as a differentiation method is invalid.
b. this particular assay method has a combinatorially large number of possible outcomes. Consequently, it is intractable to perform an unbiased assay.

1 Like
86

Get it published and peer reviewed. You first said pure Rubra was close to extinct with stern certainty, now your claim seems to be that it is impossible to tell a pure Rubra, which would mean that they may be plentiful or not but you don’t know. The two don’t go together.

1 Like
87

Typically this is done in the opposite order.

Sections (A) and (B.a) were published outside the horticultural literature 2 decades ago. Many horticulturalists ignore it; having been taken to the cleaners by hardware manufacturers.

The subject is taboo in more than one plant science journal.

True.
“At least one person in this thread has provided leaf samples to M.P. Nepal’s research program in South Dakota. The endangered nature of M. rubra is a tenet of his grant funding.”

… by morphological or genomic assay.

1 Like
88

Somehow, CHAT GPT resonates with more of a truth vibe for me. I work with dirt and trees so I have to run with what sounds right in this particular discussion. You never really answered the question of how you can claim Rubra is nearly extinct.

Here is CHAT’s response to your last list of claims.

This argument presents some legitimate concerns about the limitations of genetic techniques, especially when they are misunderstood or misapplied, but the conclusion—that current genetic methods of species identification are invalid or equivalent to phrenology—is both overstated and misleading.

Let’s break it down.

:small_blue_diamond: Species Concepts: No Single Universal Definition

You’re right that:

  • There is no single definition of species that works in all cases. The biological species concept (interbreeding populations), morphological species concept, ecological species concept, and phylogenetic species concept are all used depending on context.
  • Linnaeus’ morphological system was later complemented (not replaced) by Darwinian population thinking, and molecular tools have further expanded how we infer species boundaries.

But that doesn’t mean species themselves don’t exist—only that the boundary between them can be blurry in some cases.

:small_blue_diamond: Genetics: Not a Fallacy, But an Inference Tool

It’s true that:

  • Genetic comparisons don’t define species in a metaphysical sense. They infer relationships and boundaries based on shared ancestry, divergence, and gene flow.
  • In practice, genetic dissimilarity helps to operationalize species concepts. That’s crucial in systematics, conservation, breeding, and regulation.

The analogy to phrenology is inappropriate:

Phrenology was based on false premises and had no empirical validity.
Molecular genetics is based on well-established biochemical, statistical, and evolutionary principles. DNA barcoding, RAD-seq, whole genome sequencing, and even microsatellites consistently produce results that match and extend morphological and ecological observations.

:small_blue_diamond: PCR and Segment Length Critique

This is a valid technical point, especially in the case of SSR (microsatellite) markers:

  • Yes, SSRs measure repeat length near markers, and repeat length can vary independently of sequence.
  • Yes, identity-by-state (IBS) is not the same as identity-by-descent (IBD), and there are limits to resolution.

But:

  • This method is not used alone for species definition. It’s often used for within-species differentiation (e.g., cultivars, clones).
  • For species-level work, much higher-resolution tools are common: SNP arrays, whole genome comparisons, plastid phylogenies, RAD-seq, etc.
  • Even low-resolution markers can be statistically validated and are highly effective when used with replicates and proper models (STRUCTURE, PCA, Bayesian clustering, etc.).

:small_blue_diamond: Mathematical Set Logic: Misapplied Here

(A ≠ B) ⇒ (A = A) ∨ (A = C) ∨ (A = B ∩ C) ∨ (A = B ∪ C)

This attempts to reduce biological relationships to set algebra but ignores evolutionary context. The fallacy is not in the genetic analysis, but in treating biology as deterministic logic.

Taxonomy isn’t pure logic—it’s a probabilistic inference under constraints of incomplete knowledge. That’s why methods like coalescent theory, Bayesian phylogenetics, and maximum likelihood trees are used.

:small_blue_diamond: In Practice: Are These Methods Valid?

Yes. Here’s what they do well:

  • Identify hybrid zones and introgression (e.g., in oaks, mulberries, plums).
  • Track clonal lineages in agriculture.
  • Reveal cryptic species where morphology fails.
  • Understand domestication and gene flow history (e.g., Morus alba vs M. rubra, maize, peaches, wheat).

Are they perfect? No.
But they are:

  • Empirical,
  • Repeatable,
  • Grounded in evolutionary theory.

That’s not phrenology.

:white_check_mark: Conclusion

Genetic tools for species identification are imperfect but valid. The idea that they are “invalid” or a “fallacy” misrepresents both the science and its applications. When used critically and with awareness of their limits, they are among the most powerful tools in modern biology.

If your interlocutor has a specific technical criticism (e.g., use of low-resolution SSR markers to define species), that can be valid and is worth discussing. But to dismiss all genetic methods wholesale is ideological, not scientific.

1 Like
89

Species concepts are based on inductive, not deductive, reasoning. Applying classical logic to them is quite silly indeed.

2 Likes
90

Perhaps the most telling point is that genetics don’t correlate very well with arithmetical certainty. Said another way, if mathematical certainty is black and white, genetics is most often gray.

1 Like
91

I did not apply classical logic to species concepts.

92

I would agree that arithmetic doesn’t correlate very well with mainstream horticultural genetic analysis.

93

Yes I did.

94

If you look at mulberry trees growing in an area, it is very easy to see common patterns. For example, in Florence Alabama I can find thousands of mulberry trees growing in every possible place for a bird planted seed to take hold. Study the leaves for a few seconds and it is obvious they are not rubra and are consistent with pure alba. Here at my house, I can look at every mulberry tree within several miles and all of them have the phenotype of rubra. Zero trees in this area show alba traits. Why is this important? Because areas with alba introgression always show alba leaf types including glossy surfaces and lack of the velcro-like back surface.

Richard, your position is that rubra is extinct in the wild or nearly so. My position is that I am looking straight at trees that appear to be pure rubra. I don’t have to look far to find them. They grow beside my house, behind my house, on the slope down to my spring, along the roadside, down by the creek across the road, and in just about every imaginable place in this area. I suspect that we will find that rubra is nowhere near being extinct. Look closely at the leaf picture I posted earlier. Do you see any alba traits?

1 Like
95

No, you said what you did not do:

You still haven’t explained how you’re identifying the “near extinct” rubras beyond pointing us in the direction of unspecified “repository passports”. Who’s ID are you trusting, & what is it based on??

96

No. But check the papers of M. Nepal.

@DijonG
“At least one person in this thread has provided leaf samples to M.P. Nepal’s research program in South Dakota. The endangered nature of M. rubra is a tenet of his grant funding.”

97

Your guy Nepal is one of the authors @a_vivaldi quoted as demonstrating that M. rubra is sometimes monoecious (subdioecious aka cosexual):

And here is the first line of the abstract to Nepal et al’s “TAXONOMIC STATUS OF RED MULBERRY (MORUS RUBRA, MORACEAE) AT ITS NORTHWESTERN BOUNDARY”

“Morus rubra, commonly known as red mulberry, is rare in many areas of its native range in the United States.” (Emphasis added)

I’m having trouble accessing the full article in my phone atm, but the abstract goes on to say:

“The results from this study… provided salient features for identification of the two species.”

Which, to me, seems to imply morphology is a reliable means of differentiating rubras & albas, when one knows what features to look for.

Are you sure you’ve read these articles?

1 Like
98

Richard, here is a quote from his writings:

“Red mulberry is native to most of the eastern United States, common in the South, but becoming scarce in the northern third. Minnesota is at the northwestern edge of its range. A medium-sized tree, it typically has a crooked trunk and spreading crown. The branches are sparse and the twigs are rather thick. The heart-shaped leaves are deeply lobed on seedlings, but saplings have fewer lobes, and mature trees generally have unlobed leaves.”

Focus on that “common in the south” phrase. Red mulberry here where I live is thriving.

Also, I noted one error in his writings where he states that deeply lobed leaves are common on seedlings. Lobed leaves are common on large trees but only when in competition with other larger trees for sunlight. The leaf I posted earlier is from an older tree, estimated about 20 years old, but it is in the edge of thick woods beside the road in front of my house. That tree has plenty of large heart shaped leaves too, but a large proportion are lobed. My take is that lobed leaves are common where the leaves are in stiff competition for sunlight.

One concern I have is that mature mulberry is relatively rare. I’m being very specific with this statement because I only know of one mulberry more than 12 inches diameter in this area. Part of this may be the growth habit of mulberry where it grows near but not beside streams and as an edge tree in mixed forest.

3 Likes
99

I know almost nothing of mulberries, but I’d like to get a few answers to these questions:

  1. What are the “best” mulberries for zone 7 (or lower, unsure of the intent of “cold hardy”)?
    a) Illinois Everbearing?
    b) Silk Hope
    c) Stearns
    d) Gerardi Dwarf
    e) Black beauty
    f) Lawson Dawson
    g) Kip Parker
    h) ???
  2. What traits make them the best?
    a) Taste & texture
    b) Long fruiting period
    c) Low mature Height / Low Vigor
    d) Long berry size
    e) Disease resistance (popcorn disease)
    f) Ripening time (SWD)
    g) Prolific harvest
    h) Easy to grow
  3. What traits are in other mulberries (like Pakistan mulberry) that are missing from the cold hardy varieties?
    a) They have better taste
    b) fruiting period seems fine
    c) there exist hardy low vigor varieties it seems
    d) The biggest difference?
    e) there exists resistance already
2 Likes
100

  1. Pretty much, add kip Parker and Lawson Dawson off the top of my head, there are lots of other good ones

  2. Yes

  3. Size mostly

4 Likes